estimating the critical height of unsupported trenches in
TRANSCRIPT
Estimating the Critical Height of Unsupported Trenches in Unsaturated Soil
by
Adin Richard
B.Sc.E. (Civil Engineering), University of New Brunswick, 2016
A Thesis Submitted in Partial Fulfillment of the requirements for the Degree of
Master of Science in Engineering
In the Graduate Academic Unit of Civil Engineering
Supervisor: Won Taek Oh, Ph.D., P.Eng., Department of Civil Engineering
Examining Board: Kaveh Arjomandi, Ph.D., P.Eng., Department of Civil Engineering Othman Nasir, Ph.D., P.Eng., Department of Civil Engineering Mohsen Mohammadi, Ph.D., Department of Mechanical Engineering
This Thesis is accepted by the Dean of Graduate Studies
THE UNIVERSITY OF NEW BRUNSWICK
August 2018
© Adin A. Richard, 2018
ii
ABSTRACT
Trenching (or excavating) is integral to construction activities in the practice of
geotechnical, mining, tunnelling, and geo-environmental engineering. Many projects are
initiated by excavating trenches for infrastructure to be installed. Workers are often
required to enter trenches during the construction process, which may present serious risks.
Trench failures can result in death and/or damage to adjacent properties, therefore trenches
should be excavated with extreme precaution. The critical height (i.e. maximum depth that
can be excavated without failure) is the most important design consideration for ensuring
the stability of unsupported trenches. Excavation work is often done in unsaturated soils,
in which case the critical height of unsupported trenches can be more reliably estimated by
considering the influence of matric suction.
The purpose of this study is to investigate the influence of soil type, the matric suction
distribution, and surcharge pressure in assessing the critical height of unsupported trenches.
Extended Rankine earth pressure theory, the limit equilibrium method, and finite element
analyses were used to estimate the critical height of unsupported trenches in two
unsaturated soils.
iii
ACKNOWLEDGEMENTS
Thank you to all friends, family, and faculty for the emotional and the financial support.
There are countless individuals that made this degree possible. First of all, my parents,
Sean and Holly, you have positioned me to succeed and for that I will be forever grateful.
I am especially thankful for the number of times you have been my shuttle for kayaking.
Dad, I am sorry for the number of times us dirtbags have piled in the truck and messed it
up. Mom, thank you for continually making him take us. I genuinely could not have asked
for better parents. Stuart, my brother, I am most thankful for the number of times you
cooked me supper over the last two years. Seriously, you should open a stir-fry restaurant
(name it Stu-Fry ©). I was fortunate enough to share an office with two great colleagues,
Vitus Ileme, and Greg Brennan. I am happy to call both of you my friends. Our creative
discussions always kept me thinking. Vitus, your positive attitude and composure always
helped me on my down days. Greg, your knack for problem solving is something to admire.
Thank you both for all that you contributed to this thesis. Dr. Haralampides and Dr.
MacQuarrie, I thoroughly enjoyed each class I had with you going all the way back to
undergrad. Thank you both for helping round out my knowledge in hydraulics and
hydrology. Lastly, thank you, my supervisor, Dr. Oh. I am honoured to have been accepted
as your first ever graduate student. I think we both took a leap of faith. I did not actually
understand soil mechanics until that first semester when you taught us the ‘Mechanics of
Unsaturated Soils’. You have brought me a long way. And of course, thank you to my
friends that were always there to help me unwind. I could not have done without every one
of you!
iv
TABLE OF CONTENTS
ABSTRACT ................................................................................................................. ii
ACKNOWLEDGEMENTS ............................................................................................ iii
TABLE OF CONTENTS ................................................................................................ iv
LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES ........................................................................................................ vii
LIST OF SYMBOLS & ABBREVIATIONS ............................................................... xii
CHAPTER 1 - Introduction ............................................................................................. 1
1.1 Problem Statement ............................................................................................... 1
1.2 Objectives of the Thesis ....................................................................................... 2
1.3 Scope of the Thesis .............................................................................................. 3
1.4 Organization of the Thesis ................................................................................... 3
CHAPTER 2 – Trenching Hazards and Regulations in Canada ................................. 5
2.1 Trench Collapse Statistics .................................................................................... 6
2.2 Trench Failure Mechanisms ................................................................................. 8
2.3 Trench Failure Modes ........................................................................................ 10
2.4 Work Protection Methods .................................................................................. 12
2.5 Trench Economics .............................................................................................. 16
CHAPTER 3 – Theoretical Background ...................................................................... 18
3.1 Properties of Unsaturated Soil ............................................................................ 18
3.2 Shear Strength of Unsaturated Soil .................................................................... 23
3.3 Critical Height of Unsupported Vertical Trenches ............................................ 29
3.3.1 Rankine’s Earth Pressure Theory (1857) .................................................... 29
3.3.2 Pufahl et al. (1983) ...................................................................................... 33
3.3.3 Vanapalli & Oh (2012) ............................................................................... 36
CHAPTER 4 – Estimating the Critical Height of Unsupported Vertical Trenches in
Sand ........................................................................................................ 39
4.1 Soil Properties .................................................................................................... 40
4.2 Estimating the Critical Height with EREPT ...................................................... 44
v
4.3 Estimating the Critical Height with BSM .......................................................... 51
4.4 Comparison of Critical Heights from EREPT and BSM ................................... 58
4.5 Summary and Conclusions ................................................................................. 61
CHAPTER 5 – Estimating the Critical Height of Unsupported Trenches with
Different Wall Slopes in Sand .............................................................. 62
5.1 Estimating the Critical Height with the Finite Element Approach .................... 63
5.1.1 Hydraulic Conductivity Function ............................................................... 63
5.1.2 Analysis in SIGMA/W ................................................................................ 65
5.1.3 Analysis in SLOPE/W ................................................................................ 67
5.2 Estimating the Critical Height with the Limit Equilibrium Method .................. 70
5.3 Comparison of Limit Equilibrium and Finite Element Approaches .................. 74
5.4 Summary and Conclusions ................................................................................. 86
CHAPTER 6 – Estimating the Critical Height of Unsupported Vertical Trenches
Subjected to Surcharge Pressure ......................................................... 88
6.1 Soil Properties .................................................................................................... 88
6.2 Estimating the Critical Height without Foundation Stress ................................. 90
6.2.1 Estimating the Critical Height with EREPT ............................................... 90
6.2.2 Estimating the Critical Height with BSM ................................................... 95
6.2.3 Comparison of Critical Heights from EREPT and BSM ............................ 97
6.3 Estimating the Critical Height with Foundation Stress ...................................... 99
6.4 Summary and Conclusions ............................................................................... 107
CHAPTER 7 – General Conclusions........................................................................... 108
7.1 Recommendations for Future Research ........................................................... 111
REFERENCES .............................................................................................................. 112
CURRICULUM VITAE
vi
LIST OF TABLES
Table 2.1. Summary of the maximum allowable height of an unsupported vertical trench
............................................................................................................................................. 6
Table 2.2. Total cost breakdown for pipe-laying crew (Gulf Operators 2018) ................ 17
Table 4.1. Basic soil properties of Unimin 7030 sand (Mohamed & Vanapalli 2006) .... 41
Table 4.2. Summary of fitting parameters used to plot the SWCC of Unimin 7030 sand
........................................................................................................................................... 42
Table 4.3. Variation of the critical height in Unimin 7030 sand ...................................... 58
Table 6.1. Basic soil properties of IHT (Vanapalli et al. 1997, Oh & Vanapalli 2010) ... 89
Table 6.2. Variation of the critical height in IHT ............................................................. 97
vii
LIST OF FIGURES
Figure 2.1. Factors affecting trench stability (Ontario 2017) ............................................ 9
Figure 2.2. Excavation adjacent to existing foundations (Ontario 2017) .......................... 9
Figure 2.3. Trench parallel with backfill material (Ontario 2017) ................................... 10
Figure 2.4. Sequential trench failure (Alberta 2009) ....................................................... 11
Figure 2.5. Trench failure modes (Manitoba 2011) ......................................................... 12
Figure 2.6. Sloping standards (Ontario 2017) .................................................................. 14
Figure 2.7. Placement of spoil pile (Ontario 2017) .......................................................... 14
Figure 2.8. Shored trench with partially sloped walls (Ontario 2017) ............................. 15
Figure 2.9. Typical shored trench (Ontario 2017) ........................................................... 16
Figure 3.1. Surface tension phenomenon at air-water interface; (a) intermolecular forces
acting on contractile skin and (b) surface tension forces associated with curved 2D
surface (Fredlund et al. 2012) ........................................................................................... 19
Figure 3.2. Typical SWCC showing different zones of desaturation (Nishimura et al.
2008) ................................................................................................................................. 20
Figure 3.3. Pore-water profiles in the vadose zone (Fredlund & Rahardjo 1993a) ......... 21
Figure 3.4. Typical SWCC for silt soil (Fredlund et al. 2012) ........................................ 22
Figure 3.5. Extended Mohr-Coulomb failure envelope for unsaturated soil (Fredlund &
Rahardjo 1993b)................................................................................................................ 25
Figure 3.6. Variation of shear strength with respect to suction for four sands (modified
after Donald 1957) ............................................................................................................ 26
Figure 3.7. Relationship between and Ip (Garven & Vanapalli 2006) .......................... 28
Figure 3.8. Soil structure system for the Rankine solution for = 90° (modified after
Bowles 2001) .................................................................................................................... 30
Figure 3.9. General conditions of Mohr’s circle to derive the Rankine earth pressure
equations (modified after Bowles 2001) ........................................................................... 31
Figure 3.10. AEP and critical height in saturated soil (modified after Pufahl et al. 1983)
........................................................................................................................................... 32
viii
Figure 3.11. Lateral earth pressure states for saturated and unsaturated conditions (Pufahl
et al. 1983) ........................................................................................................................ 33
Figure 3.12. Components of AEP distribution and critical height in unsaturated soil .... 35
Figure 3.13. Unsupported vertical trench (unsaturated, intact, modified after Pufahl et al.
1983) ................................................................................................................................. 36
Figure 3.14. Field tests on the stability of an unsupported excavation in an unsaturated
soil (modified after Whenham et al. 2007) ....................................................................... 37
Figure 4.1. Measured matric suction profile and assumed hydrostatic matric suction
profile with the water table at a depth of 600-mm from the soil surface (modified after
Mohamed & Vanapalli 2006) ........................................................................................... 40
Figure 4.2. Grain size distribution curve of Unimin 7030 sand ....................................... 41
Figure 4.3. SWCC of Unimin 7030 sand (modified after Mohamed & Vanapalli 2006) 42
Figure 4.4. Direct shear test results on Unimin 7030 sand (modified after Mohamed &
Vanapalli 2006) ................................................................................................................. 43
Figure 4.5. Example of regression curve for the net AEP distribution with D = 0.6-m .. 45
Figure 4.6. Positive, negative, and net AEP distribution (D = 0.3-m) ............................. 47
Figure 4.7. Positive, negative, and net AEP distribution (D = 0.6-m) ............................. 48
Figure 4.8. Positive, negative, and net AEP distribution (D = 1.0-m) ............................. 49
Figure 4.9. Positive, negative, and net AEP distribution (D = 1.5-m) ............................. 50
Figure 4.10. Forces acting on the ith slice in Bishop’s simplified method of slices (Craig
2004) ................................................................................................................................. 51
Figure 4.11. Establishing PWP profile with In-situ analysis in SIGMA/W .................... 53
Figure 4.12. Analysis tree for Bishop’s simplified method with In-situ as the parent
analysis .............................................................................................................................. 53
Figure 4.13. Calculation of PWP based on the location of GWT in SIGMA/W ............. 54
Figure 4.14. Schematic of the ‘Entry and Exit’ slip surface method in SLOPE/W ......... 55
Figure 4.15. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 0.3-m) ....................................................................................................................... 56
Figure 4.16. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 0.6-m) ....................................................................................................................... 56
ix
Figure 4.17. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 1.0-m) ....................................................................................................................... 57
Figure 4.18. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 1.5-m) ....................................................................................................................... 57
Figure 4.19. Variation of the critical height with respect to the depth of the GWT using
extended Rankine earth pressure theory and Bishop’s simplified method (Unimin 7030
sand) .................................................................................................................................. 60
Figure 5.1. Different types of trenches considered in this chapter .................................. 63
Figure 5.2. Hydraulic conductivity function of Unimin 7030 sand ................................. 65
Figure 5.3. Boundary conditions required for coupled analyses in SIGMA/W ............... 67
Figure 5.4. Example of slope stability analysis tree used in the coupled analyses (10-s
time steps between excavations) ....................................................................................... 68
Figure 5.5. Example of slope stability analysis using SIGMA-Stress method in
SLOPE/W (1.5V:1H) ........................................................................................................ 69
Figure 5.6. Example of half-sine function used in M-P analyses (λ = 0.25) .................... 71
Figure 5.7. Forces acting on a slice within an arbitrary slip surface (GEO-SLOPE 2015)
........................................................................................................................................... 72
Figure 5.8. Variation of deformation, PWP, and FOS with time for (a) 10-s, (b) 250-s, (c)
500-s, and (d) 750-s after 1.3-m excavation stage with initial D = 0.7-m ........................ 76
Figure 5.9. FOS vs. Time for 1.3-m excavation stage (1.5V:1H) with initial D = 0.7-m 76
Figure 5.10. Variation of the critical height for different sloping angles in Unimin 7030
sand (10-s time steps, coupled analysis) ........................................................................... 78
Figure 5.11. Variation of the critical height for different sloping angles in Unimin 7030
sand (1,000-s time steps, coupled analysis) ...................................................................... 79
Figure 5.12. Variation of the critical height for different sloping angles in Unimin 7030
sand (M-P method) ........................................................................................................... 79
Figure 5.13. Variation of the critical height for 90° slope in Unimin 7030 sand ............ 80
Figure 5.14. Variation of the critical height for 3V:1H slope in Unimin 7030 sand ....... 80
Figure 5.15. Variation of the critical height for 2V:1H slope in Unimin 7030 sand ....... 81
Figure 5.16. Variation of the critical height for 1.5V:1H slope in Unimin 7030 sand .... 81
x
Figure 5.17. Distribution of total normal stress during 0.3-m excavation stage for
1.5V:1H slope with initial D = 1.2-m (10-s time steps) ................................................... 84
Figure 5.18. Distribution of shear stress during 0.3-m excavation stage for 1.5V:1H slope
with initial D = 1.2-m (10-s time steps) ............................................................................ 84
Figure 5.19. Vertical trench with D = 0.9-m (M-P method) ............................................ 85
Figure 5.20. Partially sloped (1:1) vertical trench with D = 0.9-m (M-P method) .......... 85
Figure 5.21. Variation of the critical height for 90*** slope in Unimin 7030 sand ........ 86
Figure 6.1. Grain size distribution curve of IHT (Oh &Vanapalli 2010) ........................ 89
Figure 6.2. SWCC of IHT (Oh & Vanapalli 2018) .......................................................... 90
Figure 6.3. Positive, negative, and net AEP distribution (D = 0-m) ................................ 91
Figure 6.4. Positive, negative, and net AEP distribution (D = 1-m) ................................ 92
Figure 6.5. Positive, negative, and net AEP distribution (D = 3-m) ................................ 93
Figure 6.6. Positive, negative, and net AEP distribution (D = 5-m) ................................ 94
Figure 6.7. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 0-m) .......................................................................................................................... 95
Figure 6.8. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 1-m) .......................................................................................................................... 95
Figure 6.9. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 3-m) .......................................................................................................................... 96
Figure 6.10. Slope stability analysis in SLOPE/W using Bishop’s simplified method
(D = 5-m) .......................................................................................................................... 96
Figure 6.11. Variation of the critical height with respect to the depth of the GWT using
extended Rankine earth pressure theory and Bishop’s simplified method (IHT) ............. 98
Figure 6.12. SLOPE/W model used to consider foundation stress ................................ 100
Figure 6.13. Stability analysis considering 50-kPa surcharge at a distance of 3-m
(D = 5-m) ........................................................................................................................ 101
Figure 6.14. Critical height vs. distance of foundation stress from excavation (D = 0-m)
......................................................................................................................................... 101
Figure 6.15. Critical height vs. distance of foundation stress from excavation (D = 1-m)
......................................................................................................................................... 102
xi
Figure 6.16. Critical height vs. distance of foundation stress from excavation (D = 2-m)
......................................................................................................................................... 102
Figure 6.17. Critical height vs. distance of foundation stress from excavation (D = 3-m)
......................................................................................................................................... 103
Figure 6.18. Critical height vs. distance of foundation stress from excavation (D = 5-m)
......................................................................................................................................... 103
Figure 6.19. Critical height vs. depth of the GWT with distance of foundation stress from
excavation = 0-m ............................................................................................................ 105
Figure 6.20. Critical height vs. depth of the GWT with distance of foundation stress from
excavation = 1-m ............................................................................................................ 105
Figure 6.21. Critical height vs. depth of the GWT with distance of foundation stress from
excavation = 2-m ............................................................................................................ 106
Figure 6.22. Critical height vs. depth of the GWT with distance of foundation stress from
excavation = 3-m ............................................................................................................ 106
xii
LIST OF SYMBOLS & ABBREVIATIONS
Abbreviations
AEP = active earth pressure
AEV = air-entry value
BSM = Bishop’s simplified method
CW = constant water content triaxial test
EREPT = extended Rankine earth pressure theory
FOS = factor of safety
GWT = groundwater table
IHT = Indian Head till
LEM = limit equilibrium method
M-P = Morgenstern-Price
MESA = modified effective stress approach
MTSA = modified total stress approach
PAP = pore-air pressure
PWP = pore-water pressure
SWCC = soil-water characteristic curve
Symbols
= inclination of slice base (°)
wall inclination (°)
= length of slice base (m)
backfill inclination (°)
= unit weight of soil (kN/m3)
xiii
= normalized water content
= volumetric water content
failure angle
= the first derivative of the SWCC with respect to suction
= fitting parameter for shear strength
= scaling factor for specified force function used in the Morgenstern-Price method
= Poisson’s ratio
= total stress (kPa)
= effective stress (kPa)
au = net normal stress (kPa)
= shear strength (kPa)
= internal friction angle (˚)
b = angle describing rate of change between suction and shear strength (˚)
= soil specific parameter for effective stress in unsaturated soil
= soil suction (kPa)
, , a m n = fitting parameters for Fredlund & Xing’s equation (1994)
C = total cohesion (kPa)
cwC = total cohesion from constant water content triaxial test (kPa)
c = effective cohesion (kPa)
D = depth of the groundwater table from the soil surface (m)
E = elastic modulus (kPa)
interslice normal force (kN/m)
xiv
e = Napier’s constant (2.71828…)
void ratio
( )f x = specified force function in the Morgenstern-Price method
sG specific gravity of soil particles
H = height (m)
pI = plasticity index
i = the interval between the range of j to N
j = the least negative pore-water pressure to be described by the final hydraulic conductivity function (kPa)
aK = Rankine’s active earth pressure coefficient
0K = Rankine’s earth pressure coefficient for at-rest conditions
k = hydraulic conductivity (m/s)
crm = ratio of the depth of the groundwater table to the critical height
N = slice base normal force (kN/m)
the maximum negative pore-water pressure to be described by the final hydraulic conductivity function (kPa)
ap = net active earth pressure (kPa)
, , R f x = geometric parameters as described in Figure 5.7
sR = radius of curvature of meniscus (m)
S = degree of saturation
rS = shear resistance (kN/m)
mS = mobilized shear force (kN/m)
sT = tensile force per unit length (kN/m)
au = pore-air pressure (kPa)
xv
wu = pore-water pressure (kPa)
u = pressure difference (kPa)
a wu u = matric suction (kPa)
( )a w bu u = matric suction at the air-entry value (kPa)
( )a w ru u = matric suction at the residual suction value (kPa)
W = weight of a slice within a sliding mass (kN/m)
X = interslice shear force (kN/m)
y = a dummy variable of integration representing the logarithm of negative pore-water pressure
tZ = depth of tension zone (m)
z = depth below soil surface (m)
Subscripts
FOS
f = factor of safety for force equilibrium
m = factor of safety for moment equilibrium
sat = unit weight of saturated soil
unsat = unit weight of unsaturated soil
w = unit weight of water
s = saturated water content
r = residual water content
xvi
1 = major principal stress (kPa)
3 = minor principal stress (kPa)
h = lateral pressure (kPa)
v, z = overburden pressure (kPa)
sat = shear strength of saturated soil
unsat = shear strength of unsaturated soil
H
cr = critical height of a trench
max = maximum negative pressure head
sat = height of saturated soil layer
unsat = height of unsaturated soil layer
k
sat = the measured hydraulic conductivity for saturated conditions (m/s)
unsat = the calculated hydraulic conductivity for unsaturated conditions (m/s)
1
CHAPTER 1
INTRODUCTION
1.1 Problem Statement
Most engineering projects involving foundations, landfills, pipelines, storm drains, etc., are
initiated with an excavation for infrastructure to be installed. Workers are often required to
enter the excavation or trench during the construction process, which may present serious
risks. Trenching is inherently dangerous due to the risk of cave-in, which may result in
severe injury, death, and/or consequential damage to adjacent property. Thousands of
work-related deaths and injuries in the construction industry have been attributed to trench
collapses. An average of 50 fatalities were reported each year from 1992 to 2007 in the
U.S. alone (NIOSH 2013). In Canada, each province enforces strict regulations with
respect to safe excavation practices in an attempt to prevent fatalities and serious injuries
resulting from trench collapses. The regulations specify the maximum allowable height of
an unsupported vertical trench (i.e. safe height), maximum sloping and benching angles,
and the minimum distance from the trench for stockpiling excavated or backfill materials.
Regardless of in-situ field conditions, Canadian provinces enforce safe heights in the range
of 1.2-m (New Brunswick 1991) to 1.5-m (Alberta 2009).
Trench boxes are often a practical solution in protecting workers since they allow them to
safely access the work space. However, typical trench boxes are eight metres in length and
weigh multiple tons (Trench Shoring Company 2016). Hence, the process of lifting a box,
setting it in place, and repositioning it when necessary can be very time consuming. They
2
also create an obstacle for the workers and the equipment that are used for setting
infrastructure. Lost production time results in lost profit, making it desirable to excavate
and work in unsupported vertical trenches whenever possible. However, unsupported
trenches risk collapsing and therefore must be designed and excavated with extreme
precaution, especially when workers are required to enter the excavation.
The critical height (i.e. maximum depth of a trench that can be excavated without failure,
Hcr) is the most important design consideration for ensuring the stability of unsupported
trenches. Many construction projects involve trenching and setting infrastructure in the
vadose zone; thus, the critical height of unsupported trenches should be determined by
extending the mechanics of unsaturated soils. Trench stability is mainly governed by the
matric suction distribution between the soil surface and the groundwater table (Pufahl et al.
1983, Whenham et al. 2007, De Vita et al. 2008, Vanapalli & Oh 2012). In other words,
adhering to a universal safe height suggested by provincial regulations may not be a
reasonable approach in geotechnical engineering practice, because in-situ field conditions
are not considered.
1.2 Objectives of the Thesis
The main objective of this study is to investigate the variation of the critical height of
unsupported trenches in unsaturated soil considering various field conditions and matric
suction distributions. More details are as follows:
- Review current trenching regulations and procedures in Canada.
- Review methodologies for estimating the critical height of unsupported vertical
trenches in unsaturated soil.
3
- Estimate the critical height of unsupported trenches in unsaturated cohesionless
(Unimin 7030 sand) and cohesive (Indian Head till) soils using analytical and
numerical approaches.
1.3 Scope of the Thesis
Multiple excavation scenarios were simulated considering five major factors; i) level of the
groundwater table (hereafter referred to as GWT) ii) slope of the trench wall, iii) excavation
rate, iv) magnitude of surcharge pressure, and v) distance between a trench and a surcharge
pressure. The critical heights were estimated using three distinct approaches; i) analytical
approach with extended Rankine earth pressure theory, ii) limit equilibrium methods
(Bishop’s simplified method and Morgenstern-Price method), and iii) slope stability
analyses based on finite element analyses. Geotechnical modelling software, SLOPE/W
and SIGMA/W (GEO-SLOPE International Ltd.), were used for the limit equilibrium
methods and finite element analyses. The critical heights obtained from the three different
approaches are discussed. The results were also compared to the guidelines set by Canadian
provinces. Lastly, recommendations were provided for the direction of future studies in
estimating the critical height of unsupported trenches.
1.4 Organization of the Thesis
This thesis consists of seven chapters including ‘Introduction’ (Chapter 1) and ‘General
Conclusions’ (Chapter 7).
4
Chapter 2 discusses trenching regulations and general trenching guidelines in Canada. The
modes and mechanisms of trench failure are described, and field procedures for mitigating
worker risks are discussed. Trenching economics are briefly noted.
Chapter 3 is an extensive literature review pertaining to the mechanical properties of
unsaturated soil and analytical approaches for estimating the critical height of unsupported
vertical trenches in unsaturated soil.
Chapter 4 details the procedures to estimate the critical height of unsupported vertical
trenches in an unsaturated sand using two independent methods; i) extended Rankine earth
pressure theory and ii) Bishop’s simplified method, which is a form of the Limit
Equilibrium Method (hereafter referred to as LEM). The critical heights obtained from both
approaches are compared. In addition, details of the geotechnical modelling software,
SLOPE/W and SIGMA/W, are provided.
Chapter 5 investigates the influence of excavation rate on the critical height of various
sloped trenches in an unsaturated sand through coupled stress-PWP analyses, which is a
finite element approach. The estimated critical heights for different time steps (i.e. fast and
slow excavation rates) are compared with those from the Morgenstern-Price method, which
is also a form of the LEM. Additional details of SIGMA/W are provided.
Chapter 6 examines the critical height of unsupported vertical trenches in an unsaturated
cohesive soil using the LEM (Morgenstern-Price method). The trenches being modelled
are subjected to surcharges pressures at various distances from the edge of excavation.
5
CHAPTER 2
TRENCHING HAZARDS AND REGULATIONS IN CANADA
‘Trench’ is defined as an excavation in which the measurement of depth exceeds the width.
Working in trenches is inherently dangerous because workers are subjected to the risk of
cave-ins. According to NIOSH (2013), trenches as shallow as 1.2-m deep have reportedly
collapsed and resulted in deaths. To avoid these types of tragedies, each Canadian province
enforces strict regulations with respect to safe excavation practices. Table 2.1 summarizes
the maximum allowable height of an unsupported vertical trench as regulated by each
Canadian province before safety measures such as benching, sloping, bracing, or trench
boxes must be implemented to access the workspace (i.e. safe height). The problem with
enforcing a universal safe height throughout each province is that the standards are solely
based on empirical data without any theoretical background, rather than considering the
soil type and field conditions. More importantly, the factor of safety used in determining
the safe height in each province is unknown. This may lead to over, or under conservative
protective measures being used during construction. De Vita et al. (2008) reported that
stable vertical cuts can be approximately 15-m deep in a Pozzolan deposit, while Richard
et al. (2017) concluded that the critical height in an unsaturated sand can be even less than
1-m. These studies indicate that the safe height of an unsupported trench should be
estimated based on the job site conditions.
6
Table 2.1. Summary of the maximum allowable height of an unsupported vertical trench
Provinces in Canada Safe height
BC, NB, NS, ON, QC, SK (NL, PE) 1.2-m (4-ft)
AB, MB 1.5-m
2.1 Trench Collapse Statistics
According to NIOSH (2013), an average of 50 fatalities were reported annually in the U.S.
from 1992-2007 due to trench collapses. Eivemark & Hall (2000) stated that approximately
1% of work-related deaths in the U.S. were caused by trenching accidents. Other studies
showed that approximately 1,000 injuries, and between 60 and 100 deaths per year were
associated with trench failures in the U.S. (Thompson & Tanenbaum 1977, Suruda et al.
1988, White 2008). Most deaths occurred in sewer line construction when the trenches
were not shored. 80% of reported fatalities occurred in trenches less than 4.5-m deep, and
40% occurred in trenches less than 3-m deep (Eivemark & Hall 2000). Shoring or trench
boxes normally provide sufficient worker protection. However, workers should exit the
trench before shoring or a trench box is moved (Hinze 2005).
The number of deaths and injuries related to trenching have significantly declined as
regulations have been improved. In New Brunswick, the last death that was caused by
trenching was reported on May 28, 2001 (WorkSafeNB 2017). The employer was charged
with non-compliance of regulations 181(1) and 182(1)(a) of the Occupational Health &
Safety Act (1991), which are listed below:
7
181(1): An employer shall ensure that the walls of an excavation or trench are
supported by shoring, bracing, or caging, except when the excavation or trench
(a) is less than 1.2-m deep,
(b) subject to subsection (2), is cut in solid rock,
(c) is sloped or benched to within 1.2-m of the bottom of the excavation or trench
with the slope or bench not exceeding 1-m of vertical rise to each 1-m of
horizontal run, or
(d) is one that an employee is not required to enter.
181(2): Where the walls or crests of an excavation or trench are cut in solid rock
and are not stable, an employer shall ensure that the walls and crests are adequately
supported by rock bolts, wire mesh, shoring, or a method that provides equivalent support.
182(1): An employer shall ensure that an employee does not, and no employee shall,
enter an excavation or trench 1.2-m or more in depth unless
(a) the walls of the excavation or trench are supported by shoring, bracing, or
caging, the excavation or trench is cut in solid rock, or the excavation or trench
is sloped or benched to within 1.2-m of the bottom of the excavation or trench
with the slope not exceeding 1-m of vertical rise to each 1-m of horizontal run.
Only infraction codes were provided in the investigation report without site-specific details.
The employer was fined $4,800 for their negligence.
8
2.2 Trench Failure Mechanisms
Changes in pore-water pressure (hereafter referred to as PWP), surface loading, and
vibration are the most predominant causes of trench instability (Ontario 2017). Figure 2.1
illustrates the causes of trench failure.
The base and walls of a trench experience elastic rebound immediately after excavating
due to the relief of confining pressures. This causes an increase in the soil’s void space and
a decrease in PWP, thus increasing the soil’s shear strength. Hence, a trench may appear
stable immediately after excavating. However, the trench destabilizes over time as PWP
rises back to equilibrium within the soil nearby the trench base and walls. Rainfall
infiltration, development of cracks, and a rising GWT can further increase PWP and
accelerate a trench failure. Thus, the risk of trench collapse increases as exposure time to
the atmosphere increases. Trenches should be backfilled as soon as the job is complete.
Many forms of static and dynamic loading can cause a trench to collapse. Surcharge loads
from excavated material and construction equipment, and vibration induced by vehicular
traffic, compaction, pile driving, blasting, other adjacent excavations, etc. may trigger a
collapse. Trenches excavated near existing foundations may collapse if the bearing
pressure enters the active zone of the trench (Figure 2.2).
9
Figure 2.1. Factors affecting trench stability (Ontario 2017)
Figure 2.2. Excavation adjacent to existing foundations (Ontario 2017)
The presence of buried utilities adjacent to an open excavation can result in an unexpected
failure because the backfill materials encasing the utility are typically weaker than
undisturbed native soil (Figure 2.3). This type of failure mechanism not only endangers
workers, but may also result in excessive strain causing structural damage to the utilities
10
(Symons et al. 1982). Shoring is typically required when excavating near buried utilities or
foundation structures to ensure ground movement is minimized.
Figure 2.3. Trench parallel with backfill material (Ontario 2017)
Heterogeneity should also be monitored when excavating, especially over long distances.
Some soils such as gravel deposits may have a high variance with respect to local
mechanical properties (Krahn & Fredlund 1983, Dai et al. 2016). Thus, the critical height
may vary significantly between locations in the same soil deposit. In such cases, it may be
necessary to perform a probabilistic analysis to determine minimum values for the soil
properties of interest to quantify a critical height.
2.3 Trench Failure Modes
The Occupational Health & Safety Division of Alberta (2009) described how rescue
attempts may be more difficult and dangerous when the wall failure occurs sequentially as
shown in Figure 2.4. In this case, failure is initiated at the base of the trench wall. Cracking
near the ground surface and local failures in Zone 1 should raise alarms. This localized
failure or movement leads to failure in Zone 2. Finally, the failure in Zone 3 occurs due to
the self-weight of the soil. This failure sequence is a plausible explanation for why rescuers
11
are sometimes trapped along with the first victim(s). Someone attempting to intervene and
help uncover a victim when failure in Zone 3 has not yet occurred can put themselves at
risk.
Figure 2.4. Sequential trench failure (Alberta 2009)
The Workplace Safety & Health Division of Manitoba (2011) categorizes the modes of
trench collapse into four types (Figure 2.5).
(a) Spoil pile slide - occurs when the excavated material is not placed far enough
away from the edge of the excavation. A minimum distance of 0.6-m is
recommended for every one-metre of excavation depth.
(b) Side wall shear - common to fissured or desiccated clay-type or alluvial soils that
are exposed to drying.
(c) Slough-in (cave-in) - common to previously excavated material, fill, and granular
soils where the water table is above the base of the excavation, or where soils are
organic or peat.
(d) Rotation - common in clay-type soils when excavation walls are too steep, or
when moisture content increases rapidly.
12
Figure 2.5. Trench failure modes (Manitoba 2011)
2.4 Work Protection Methods
At least one of benching, sloping, shoring, or worker-protective trench boxes must be
implemented in the trenching process when it is necessary to excavate deeper than the
regulated safe height. The preferred safety measure depends on the application and the soil
properties. Protection methods have been devised based on four soil types (Figure 2.6;
Ontario 2017):
(a) Type 1 - includes hardpan, consolidated clay, and some glacial tills. It is hard to
drive a pick into Type 1 soil. The sides remain vertical with no water released
from the trench wall during excavation. Rain may cause soil to break down along
the edges of the excavation.
(a) (b)
(c) (d)
13
(b) Type 2 - Includes silty clay and less dense tills. It is relatively easy to drive a pick
into Type 2 soil and can be easily excavated with a backhoe. The sides will
remain vertical for short periods of time before tension cracks appear as the soil
starts to dry.
(c) Type 3 - Includes sand, granular materials, silty or wet clays, and all backfill or
previously disturbed material. Type 3 soil can be excavated without difficulty
using a backhoe and will not stand vertically when dry. Soil will stand vertically
for short periods of time when wet, however it dries quickly and is prone to
deterioration by vibration from heavy equipment.
(d) Type 4 - Includes quicksand, Leta clays, silty clays, muskeg, or other organic
deposits with high moisture content. Type 4 soil is flowable material that can be
excavated with no difficulty and must be supported or contained to excavate to
any significant depth.
Regardless of the material, the spoil pile should be placed further away from the edge of
excavation as depth increases (Figure 2.7; Ontario 2017). It is necessary to slope the trench
walls 1:1 once the excavation extends deeper than the safe height in Type 1 and 2 soils.
Type 3 and 4 soils warrant 1:1 and 1V:3H slopes from the base of the excavation,
respectively (Figure 2.6). The main restriction with sloping is the space allotted to achieve
an adequate slope, and therefore other reinforcement may be required in certain practical
scenarios. Figure 2.8 illustrates the case of using shoring in conjunction with sloping.
14
Figure 2.6. Sloping standards (Ontario 2017)
Figure 2.7. Placement of spoil pile (Ontario 2017)
Good soil
Faily good soil Type 3 soil
Maximum1.2-m
Bad soil Type 4 soil
Type 1 & 2 soil
Minimumbank slope
Minimumbank slope
Minimumbank slope
15
Figure 2.8. Shored trench with partially sloped walls (Ontario 2017)
Shoring involves supporting trench walls with a braced structure as shown in Figure 2.9.
Shoring systems are used to transfer and resist the loads acting on opposing sides of the
trench. Hydraulic shoring is more frequently used in practice than timber shoring (Hinze
2005). Hydraulic shores are prefabricated sections in which the width can be adjusted with
a pneumatic device, making them much more practical than timber shores that must be cut
to the exact width of the excavation. Trench boxes are different from shores as they do not
provide structural support to trenches unless the void space between the box and the walls
are backfilled; their sole purpose is to prevent soil from collapsing onto the workers. For
this reason, trench boxes should always be accessed by a ladder which rests inside the box.
The space between the box and the wall should be minimized to provide easier access to
the ladder from the edge of the excavation, and to limit soil movement in the event of a
collapse. It is often necessary to use shoring rather than trench boxes while working in the
proximity of infrastructure such as buried utilities, roads with nearby traffic, or buildings.
16
Figure 2.9. Typical shored trench (Ontario 2017)
2.5 Trench Economics
Unsupported trenches require the least amount of excavation work and are therefore the
most desirable form of trenches from an economical perspective. Benching and sloping
require more time for excavating since more soil is removed per linear distance, yet the
installation process may be unimpeded if the infrastructure cannot be installed as fast as
the trench is being excavated. However, production is inevitably slowed when using
shoring or trench boxes. An anonymous construction company was capable of setting 6 to
8 (0.6-m diameter, 6-m long) pipes per day when working in and around a trench box,
compared to 12 to 15 per day when working without one. Inconsistencies in day-to-day
operations were mostly attributed to human resources, and trench dewatering associated
17
with groundwater seepage or rainfall. The opportunity cost of employing an entire pipe-
laying crew with heavy equipment for a 12-hour day is approximately $10,000. A total cost
breakdown is shown below in Table 2.2. Other costs that are not listed in Table 2.2 include
the cost of quality control inspectors, surveyors, and rental equipment. Over an entire
construction season on a large-scale project, hundreds of thousands of dollars could be
saved if working without a trench box or shoring could be justified.
Table 2.2. Total cost breakdown for pipe-laying crew (Gulf Operators 2018)
Equipment Cost per hour ($CAD/hr)
Quantity Cost per 12-hour day ($CAD/day)
Cat 345 Excavator 215.00 1 2,580.00
Cat 336 Excavator 190.00 1 2,280.00
Cat D-4 Dozer 95.00 1 1,140.00
Compactor 85.00 1 1,020.00
Labour 37.00 4 1,776.00
Lead Hand 41.00 1 492.00
Supervisor 68.00 1 816.00
Total cost per day 10,104.00
18
CHAPTER 3
THEORETICAL BACKGROUND
3.1 Properties of Unsaturated Soil
Unsaturated soil consists of soil particles, water, and air (i.e. solid, liquid, and gas phases).
The three phases interact to create the contractile skin (or air-water interface), which is
referred to as the fourth phase. As shown in Figure 3.1(a), a molecule of water within a
body experiences equal forces in all directions, while a molecule along the air-water
interface experiences an unbalanced force towards the interior of the water body. Thus, a
tensile force is generated tangentially to the air-water interface to satisfy static equilibrium.
Surface tension causes the contractile skin to behave like an elastic membrane; somewhat
like an inflated balloon that has a greater air pressure inside than in the atmosphere. The
membrane forms a concave curvature towards the larger pressure (Figure 3.1(b)). The
radius of curvature is directly related to the pressure difference as shown in Eq. (3.1). The
tensile force is temperature dependant. In an unsaturated soil, the air pressure (ua = 0-gauge
pressure) becomes greater than the water pressure (uw = negative gauge pressure) which
creates a contractile skin. The pressure difference between ua and uw is referred to as matric
suction (Eq. (3.2)).
2 s
s
Tu
R (3.1)
19
2 sa w
s
Tu u
R (3.2)
where Δu = pressure difference, Ts = tensile force per unit length, Rs = radius of curvature,
and (ua – uw) = matric suction
Figure 3.1. Surface tension phenomenon at air-water interface; (a) intermolecular forces acting on contractile skin and (b) surface tension forces associated with curved 2D surface (Fredlund et al. 2012)
Eq. (3.1) shows that matric suction approaches infinity as the radius of curvature
approaches zero, indicating that smaller radii can exert greater matric suction than larger
radii. In terms of the radii within the void space of an unsaturated soil, fine-grained soils
with small pore-sizes have a higher affinity for matric suction than coarse-grained soils
(a)
(b)
20
with relatively large pores. In other words, the ability for a soil to retain water is strongly
related to the pore-size (or particle-size) distribution.
The soil-water characteristic curve (hereafter referred to as SWCC) is a soil specific
relationship between matric suction and the degree of saturation (or volumetric water
content). A typical SWCC can be divided into three main zones; boundary effect zone,
transition zone, and residual zone (Figure 3.2).
Figure 3.2. Typical SWCC showing different zones of desaturation (Nishimura et al. 2008)
The boundary effect zone ranges from zero matric suction (i.e. saturation) up to the air-
entry value (hereafter referred to as AEV). The AEV is the suction required to force air
into the largest voids in a soil and is the first point of importance on a SWCC for a drying
Suction (kPa)
1 10 100 1000 10000 100000 1000000
Deg
ree
of
satu
rati
on
, S (
%)
0
20
40
60
80
100
Boundaryeffect zone
Transitionzone
Residual zoneof
unsaturation
Air-entry value(ua - uw)b
Residualsuction value
Air
Soil
Water film
21
cycle. Soil begins desaturating once matric suction is increased past the AEV. Most
desaturation occurs in the transition zone. The degree of saturation (or water content) is
reduced significantly with relatively small increases in matric suction. In the residual zone,
large increases in suction result in small changes in the degree of saturation.
Figure 3.3 demonstrates possible PWP profiles with respect to depth above and below the
GWT. In the saturated zone, the pore-water is in compression (i.e. zero matric suction) and
the magnitude of pressure increases linearly with depth. In the unsaturated zone (or vadose
zone), the pore-water is in tension and the soil is subjected to a matric suction distribution
that can vary linearly or nonlinearly above the GWT (Fredlund & Rahardjo 1993a). Ideal
hydrostatic conditions (i.e. linear matric suction distribution; ① in Figure 3.3) often do not
represent true field conditions due to environmental factors such as evaporation (② in
Figure 3.3) and rainfall infiltration (③ in Figure 3.3).
Figure 3.3. Pore-water profiles in the vadose zone (Fredlund & Rahardjo 1993a)
22
Soils in nature experience cycles of wetting and drying (i.e. rainfall infiltration and
evaporation), which alter the SWCC of the soil accordingly. Increasing or decreasing a
soil’s moisture content by the same volume of water does not cause the same change in
suction. This phenomenon is known as hysteresis. The first drying (desorption curve) and
wetting (adsorption curve) cycle form the two boundaries for the possible SWCCs that a
soil can experience (Figure 3.4). Larger pores drain first due to better interconnectivity,
while smaller pores are more reluctant to release water when drying since there is more
surface contact between the available water and the soil grains. Conversely, larger pores
are filled last during wetting because smaller pores absorb water more easily due to a higher
affinity for matric suction. Thus, the drying SWCC always corresponds to higher suction
than the wetting SWCC for a given moisture content (Yang et al. 2004, Goh et al. 2014).
It is therefore more conservative to analyze the mechanical behaviours of unsaturated soils
using a wetting-cycle SWCC.
Figure 3.4. Typical SWCC for silt soil (Fredlund et al. 2012)
1,000,000
23
3.2 Shear Strength of Unsaturated Soil
The effective stress equation proposed by Terzaghi (1936) laid a foundation towards
establishing geotechnical engineering as one of the independent fields of civil engineering
(Vanapalli 2009). Effective stress is the difference between total stress and PWP (Eq. (3.3)).
In conjunction with Mohr-Coulomb theory, effective stress can be used to interpret the
shear strength of saturated soils (Eq. (3.4)).
wu (3.3)
tansat c (3.4)
where ’ = effective stress, = total stress, uw = pore-water pressure, sat = shear strength
of saturated soil, c’ = effective cohesion, and ’ = effective internal friction angle
Most soils, especially in arid or semi-arid regions, exist in unsaturated conditions. In this
case, the influence of matric suction should be considered to reliably estimate the shear
strength of unsaturated soil. Bishop (1959) was one of the first researchers to recognize
that suction contributes to the strength of unsaturated soil and proposed Eq. (3.5) by
extending Terzaghi’s equation for effective stress (i.e. Eq. (3.3)).
a a wu u u (3.5)
where ua = pore-air pressure, ( – ua) = net normal stress, and = a soil parameter related
to degree of saturation (or matric suction) ranging from 0 to 1
24
The effective stress parameter, χ, is equal to zero for dry soil. When the soil is saturated, χ
is equal to one and Eq. (3.5) becomes equal to Eq. (3.3). Jennings & Burland (1962) and
Bishop & Blight (1963) pointed out that there are limitations to Eq. (3.5) since different χ
values are required to explain shear strength and volume change behaviour of unsaturated
soils. In other words, Eq. (3.5) cannot universally describe the mechanical behaviours of
unsaturated soil. The parameter, , can be evaluated using Eq. (3.6) to determine the shear
strength of unsaturated soils depending on the range of matric suction values (Khalili &
Khabbaz 1998).
1
a wa w a w b
a w b
a w a w b
u uif u u u u
u u
if u u u u
(3.6)
where (ua – uw)b = air-entry value of a soil
Fredlund & Morgenstern (1977) concluded that net normal stress and matric suction are
the two independent stress-state variables that are to be used for evaluating the mechanical
behaviour of unsaturated soil. Fredlund et al. (1978) proposed Eq. (3.7) to estimate the
shear strength of unsaturated soil based on the extended Mohr-Coulomb failure envelope
(Figure 3.5).
tan tan bunsat a a wc u u u (3.7)
where unsat = shear strength of unsaturated soilb = angle indicating the rate of increase in
shear strength with respect to a change in matric suction
25
Figure 3.5. Extended Mohr-Coulomb failure envelope for unsaturated soil (Fredlund & Rahardjo 1993b)
However, determining b is time-consuming and requires elaborate testing equipment.
Previous research also showed that b is not a constant, but varies nonlinearly with respect
to matric suction (Escario & Saez 1987, Gan et al. 1988). For matric suction values less
than the AEV, b = ’, and b becomes less than ’ once matric suction exceeds the AEV
(i.e. b < ’). It is interesting to note that shear strength approaches that of saturated
condition for suction values greater than residual suction in coarse-grained soils (Figure
3.6). This is attributed to a decrease in the net contribution of matric suction towards shear
strength as the residual condition is approached. The contractile skin becomes
discontinuous in the residual stage and becomes incapable of maintaining peak strength.
26
Figure 3.6. Variation of shear strength with respect to suction for four sands (modified after Donald 1957)
Vanapalli et al. (1996) proposed equations to estimate the variation of b with respect to
suction in terms of volumetric water content/degree of saturation as shown in Eq. (3.8) and
(3.9), respectively.
tan tanb r
s r
(3.8)
tan tan100
b r
r
S S
S
(3.9)
where S = degree of saturation, = volumetric water content, subscript s = saturated
condition, and subscript r = residual condition
Matric suction, (ua - uw) (kPa)
0 5 10 15 20 25 30 35
Sh
ear
stre
ng
th (
kPa)
5
10
15
20
25
Fine FrankstonGraded FrankstonMedium FrankstonBrown Frankston
Donald 1957
27
The SWCC for the entire range of suction is required to accurately identify the volumetric
water content or degree of saturation for a soil’s residual condition. Vanapalli et al. (1996)
proposed another equation that requires the degree of saturation (or volumetric water
content) along with a fitting parameter.
tan tan tanb S (3.10)
where = normalized volumetric water content (= /s) and = fitting parameter for the
shear strength of an unsaturated soil
Vanapalli & Fredlund (2000) provided a relationship between the fitting parameter, κ, and
the plasticity index, Ip, as shown in Eq. (3.11) using five data sets of shear strength of
unsaturated soils. Garven & Vanapalli (2006) further improved Eq. (3.11) by using ten data
sets of shear strength for compacted soils (Eq. (3.12) and Figure 3.7). Eq. (3.11) and (3.12)
show that κ = 1 for estimating the variation of shear strength of unsaturated non-plastic
soils with respect to matric suction.
20.0008( ) 0.0801( ) 1p pI I (3.11)
20.0016( ) 0.0975( ) 1p pI I (3.12)
where Ip = plasticity index
28
Figure 3.7. Relationship between and Ip (Garven & Vanapalli 2006)
Total cohesion is the sum of effective and apparent cohesion as shown in Eq. (3.13).
tan ba wC c u u (3.13)
where C = total cohesion c’ = effective cohesion, and (ua uw) tanϕb = apparent cohesion
Substituting Eq. (3.10) into Eq. (3.13), and then into Eq. (3.7) yields Eq. (3.14).
tan tan
tan tan
tan tan
tan tan
tan
bunsat a a w
a a w
a w a
a w a
a
c u u u
c u u u S
c u u S u
c u u u
C u
(3.14)
Plasticity index, Ip
0 10 20 30 40 50
Fit
tin
g p
aram
eter
,
0
1
2
3
4
5
6
7
Best fit parameter68% confidence interval
Regression
= -0.0016 (Ip2)+0.0975 (Ip)+1
A. Red Silty ClayB. Madrid Gray ClayC. Adams ClayD. Indian Head Till (dry of opt.)E. Indian Head Till (opt.)
F. Botkin SiltG. Ste. Rosalie ClayH. LD Dhanauri ClayI. HD Dhanauri ClayK. Nanyang Clay
FA
C
G
D,EBIK
H
29
3.3 Critical Height of Unsupported Vertical Trenches
3.3.1 Rankine’s Earth Pressure Theory (1857)
Rankine (1857) investigated the stress conditions of a soil in a state of plastic equilibrium.
Bowles (2001) summarized the major assumptions made in Rankine’s theory as listed
below:
(a) Soil is isotropic and homogeneous and has internal friction but no cohesion.
(b) The rupture surface (AC in Figure 3.8) is a plane surface and the backfill surface
(BC in Figure 3.8) is planar (it may slope but is not irregularly shaped).
(c) The frictional resistance is distributed uniformly along the rupture surface and
the soil-to-soil friction coefficient, f = tan.
(d) The failure wedge (ABC in Figure 3.8) is a rigid body undergoing translation.
(e) There is no wall friction.
(f) Failure is a plane strain problem - that is, consider a unit interior slice from an
infinitely long wall.
30
Figure 3.8. Soil structure system for the Rankine solution for = 90° (modified after Bowles 2001)
Plastic equilibrium is achieved if a wall moves sufficiently far away from a soil mass, in
which the effective lateral earth pressure acting on the vertical plane is denoted as
Rankine’s active earth pressure (hereafter referred to as AEP). For the case shown in Figure
3.9, AEP can be calculated using Eq. (3.15) assuming the soil’s unit weight is constant.
2 2
2 2
cos
cos cos coscos
cos cos cos
a a a
a
p z K zK
K
(3.15)
where pa = active earth pressure, γ = unit weight of soil, z = depth, β = angle of backfill,
and Ka = active earth pressure coefficient
β
θ‐β
θ
90° - θ
90° + β
ϕH
B
A
W
R
C
Pa
31
(a) General case: only for +β as shown (b) Mohr’s circle
Figure 3.9. General conditions of Mohr’s circle to derive the Rankine earth pressure equations (modified after Bowles 2001)
When the backfill surface is horizontal, the AEP coefficient can be calculated with Eq.
(3.16).
1 sin
1 sinaK
(3.16)
Rankine did not incorporate soil cohesion in estimating AEP. This issue is resolved by
combining Eq. (3.17) with Rankine’s theory as shown in Eq. (3.18).
23 1 tan 45 2 tan 45
2 2c
(3.17)
2a a ap zK c K (3.18)
where 3 = minor principal stress, 1 = major principal stress
β
β
τ
σ
σv
pa O
G
Eϕσv OG
pa OE
90° β 90° β
32
The lateral earth pressure (positive, negative, and net) distribution in a saturated cohesive
soil is shown in Figure 3.10. From the ground surface to a depth of zero net AEP is referred
to as the tension zone. The depth of the tension zone can be calculated using Eq. (3.19).
Theoretically, by setting the sum of horizontal forces equal to zero, a vertical trench can be
excavated up to twice the depth of the tension zone without failure if effective cohesion
and the unit weight are constant. In which case, the critical height of an unsupported
vertical trench is calculated using Eq. (3.20).
Figure 3.10. AEP and critical height in saturated soil (modified after Pufahl et al. 1983)
2
t
a
cZ
K
(3.19)
4
2cr t
a
cH Z
K
(3.20)
Where Hcr = critical height, Zt = depth of tension zone
H
zt
+ =
z
K -2c'K v haA
33
By assuming a planar rupture surface and a horizontal backfill surface, the AEP in an
unsupported trench in unsaturated soil can be interpreted by extending Rankine’s earth
pressure theory.
3.3.2 Pufahl et al. (1983)
Pufahl et al. (1983) investigated lateral earth pressure (i.e. active and passive pressures) for
a vertical trench extending the mechanics of unsaturated soil. Conventional Rankine earth
pressure theory for saturated soil was modified to incorporate the influence of matric
suction using the Mohr-Coulomb failure criteria (i.e. extended Rankine earth pressure
theory). The lateral earth pressures (i.e. active and passive) for both saturated and
unsaturated conditions are illustrated in Figure 3.11.
Figure 3.11. Lateral earth pressure states for saturated and unsaturated conditions (Pufahl et al. 1983)
34
The AEP decreases as a soil desaturates due to increasing contribution from matric suction.
From the geometrics of the Mohr circle in Figure 3.11, the net AEP can be written as Eq.
(3.21) by replacing effective cohesion in Eq. (3.18) with total cohesion (Eq. (3.13)) to
incorporate the influence of matric suction.
2 tan bh a v a a a w au u K c u u K (3.21)
where (h – ua) = net lateral pressure, and (v – ua) = net overburden pressure
If the pore-air pressure (hereafter referred to as PAP) is assumed to be atmospheric pressure
(i.e. ua = 0), the AEP in an unsaturated soil can be calculated using Eq. (3.22).
2 tan ba v a a w ap K c u u K (3.22)
Substituting Eq. (3.10) into Eq. (3.22) yields Eq. (3.23).
2 tan
2 tan
2
ba v a a w a
v a a w a
v a a
p K c u u K
K c u u S K
K C K
(3.23)
Typically, matric suction varies nonlinearly with depth. In other words, the critical height
in unsaturated soils may not be simply two times the depth of the tension zone. In this case,
the critical height can be estimated by locating the depth that equates ‘Area I’ and ‘Area II’
as shown in Figure 3.12.
35
Figure 3.12. Components of AEP distribution and critical height in unsaturated soil
By assuming a planar failure surface, the critical height in unsaturated soil can also be
determined by considering a balance of forces on the sliding wedge shown in Figure 3.13.
Trial and error must be used to solve for Hcr since it appears on both sides of Eq. (3.24).
2
4tan
2
2 tan1 2
bcra w
acr
wcr sat cr cr unsat
a
mc u u
KH
m m mK
(3.24)
where mcr = D/Hcr, D = depth of the GWT from the soil surface, w = unit weight of water,
and sat, unsat = unit weight of soil for saturated and unsaturated condition, respectively
36
Figure 3.13. Unsupported vertical trench (unsaturated, intact, modified after Pufahl et al. 1983)
3.3.3 Vanapalli & Oh (2012)
Vanapalli & Oh (2012) extended the work by Pufahl et al. (1983) to analyze the stability
of an unsupported vertical trench in an unsaturated soil. The excavation dimensions were
3-m deep, 6-m wide, and 20-m long. The variation of matric suction with depth was
measured using tensiometers installed at depths of 1, 1.5, 2.5, and 3.5-m from the ground
surface (Whenham et al. 2007). The trench was first excavated in June 2004. The first
localized and generalized failures were observed in January 2005 and February 2005,
respectively, due to a decrease in matric suction associated with precipitation activity
(Figure 3.14).
crH
sat
unsat
37
Figure 3.14. Field tests on the stability of an unsupported excavation in an unsaturated soil (modified after Whenham et al. 2007)
Two different approaches were used to calculate AEP; the Modified Effective Stress
Approach (hereafter referred to as MESA; Eq. (3.22)) and the Modified Total Stress
Approach (hereafter referred to as MTSA; Eq. (3.25)). The MTSA implies that excess PAP
is assumed to drain and the excess PWP is an undrained condition. Therefore, AEP is
estimated by replacing total cohesion, C, in Eq. (3.22) with total cohesion from the
constant-water content triaxial test (Ccw; Rahardjo et al. 2004, Thu et al. 2006).
2a v a cw ap K C K (3.25)
The factor of safety (hereafter referred to as FOS) estimated using the MESA and the
MTSA were 0.31 and 1.14, respectively, at the time when first localized failures occurred.
38
This implies that the stability analysis using the MESA is more conservative, but the
analysis done with constant-water content test results provides more realistic estimates
compared to the field observations. At the time general failures occurred, matric suction
remained relatively constant at depth and the wetting front remained stable even after a
small rainfall event. This indicates that general failure of the excavation can be attributed
to an increase in the AEP associated with an increase in the soil’s unit weight.
In this study, Eq. (3.23) was used to calculate the net AEP distribution for an unsupported
vertical trench in unsaturated soil.
39
CHAPTER 4
ESTIMATING THE CRITICAL HEIGHT OF UNSUPPORTED VERTICAL TRENCHES IN SAND
In this chapter, an attempt is made to estimate the critical height of an unsupported vertical
trench in an unsaturated sand. Two independent approaches were used; i) extended
Rankine’s earth pressure theory (hereafter referred to as EREPT) and ii) Bishop’s
simplified method (hereafter referred to as BSM) with geotechnical modelling software,
SLOPE/W. It was assumed that the trenches were excavated into Unimin 7030 sand with
various levels of the GWT (i.e. different matric suction distributions).
Mohamed & Vanapalli (2006) conducted model-footing tests in a sand (Unimin 7030) for
both saturated and unsaturated conditions in a specially designed soil tank. Unsaturated
conditions were achieved by setting a water level at some depth, and the matric suction
distribution was established by measuring matric suction at various depths using
conventional tensiometers. Figure 4.1 shows the measured and the assumed matric suction
distributions with the water table at a depth of 600-mm in the soil tank, in which there is a
negligible difference. Hence, a hydrostatic matric suction distribution was assumed for
both approaches used in this chapter.
40
Figure 4.1. Measured matric suction profile and assumed hydrostatic matric suction profile with the water table at a depth of 600-mm from the soil surface (modified after Mohamed & Vanapalli 2006)
4.1 Soil Properties
Basic soil properties of Unimin 7030 sand are summarized in Table 4.1. The grain size
distribution curve of the sand is shown in Figure 4.2. The SWCC was measured using a
Tempe Cell apparatus extending the axis-translation technique (Figure 4.3; Hilf 1956). The
AEV [(ua – uw)b] and residual suction value [(ua – uw)r] were estimated to be 4 kPa and 7.8
kPa, respectively. This was done using the procedure detailed in Vanapalli et al. (1999),
(i.e. the matric suction values corresponding to the intersection of the two linear slope
segments of the SWCC; Figure 4.3). A best-fit analysis for the SWCC was conducted using
Fredlund & Xing's (1994) model (Eq. (4.1)), and the fitting parameters are summarized in
Table 4.2.
10
150
300
600
500
B
1 kPa
2 kPa
4 kPa
7 kPa
Tensiometer
GWT
distribution ofmeasured matric suction
mm
mm
mm
mm
mmGWT:groundwater table
hydrostatic matric suctiondistribution
41
Table 4.1. Basic soil properties of Unimin 7030 sand (Mohamed & Vanapalli 2006)
Properties Value
Plasticity Index, Ip NP
Saturated unit weight, sat (kN/m3) 20.4
Saturated water content, θs (%) 38.7
Void ratio, e 0.63
Specific gravity, Gs 2.65
Effective cohesion, c’ (kPa) 0
Effective internal friction angle, ’ (˚) 36.2
Saturated hydraulic conductivity, ksat (m/s) 5 × 10-5
Elastic modulus, E (kPa)** 10,000
Poisson’s ratio, ν ** 0.33
** = assumed value
Figure 4.2. Grain size distribution curve of Unimin 7030 sand
Particle Size (mm)
0.01 0.1 1 10
% F
iner
0
20
40
60
80
100
Sand
42
Figure 4.3. SWCC of Unimin 7030 sand (modified after Mohamed & Vanapalli 2006)
1
ln
m
ns
Se a
(4.1)
where e = Napier’s constant (2.71828…), ψ = soil suction, and a, n, m = fitting parameters
Table 4.2. Summary of fitting parameters used to plot the SWCC of Unimin 7030 sand
Parameter a m n
Value 11.415 54.202 5.1322
Soil suction, (kPa)
0 2 4 6 8 10
Vo
lum
etri
c w
ater
co
nte
nt,
0.0
0.1
0.2
0.3
0.4
0.5
Fredlund and Xing (1994)Tempe cell
(ua - uw)b = 4 kPa
(ua - uw)r = 7.8 kPa
Residual zoneBoundary effect zone Transition zone
43
Volume changes can be neglected in cohesionless soils during drying/wetting cycles (i.e.
void ratio, e = constant); therefore, the unit weight of the sand is a function of volumetric
water content as shown in Eq. (4.2). The variation of the unit weight with respect to
volumetric water content (or matric suction) was accounted for in both approaches
discussed in this chapter.
1
1s
w
G e
e
(4.2)
where Gs = specific gravity, and e = void ratio
Effective cohesion and the internal friction angle of the sand were determined from the
direct shear test (Figure 4.4).
Figure 4.4. Direct shear test results on Unimin 7030 sand (modified after Mohamed & Vanapalli 2006)
Normal stress, n (kPa)
0 50 100 150
Sh
ear
stre
ng
th,
(kP
a)
0
50
100
150
Unimin 7030 sand
c' = 0 kPa; ' = 36.2
44
4.2 Estimating the Critical Height with EREPT
The procedure for estimating the critical height of an unsupported vertical trench using
extended Rankine earth pressure theory is as follows:
(a) Set a desired elevation of the GWT.
(b) Set the increment size (i.e. ΔH) for calculating pressures. ΔH = 1-mm was used
in this study for calculating the AEP distribution.
(c) Establish the matric suction distribution with depth by multiplying the unit
weight of water by the distance from the GWT [(ua - uw) = 0 in the saturated zone].
(d) Calculate the distribution of volumetric water content with the computed values
of matric suction by using Eq. (4.1) and the values in Table 4.2.
(e) Calculate the distribution of the unit weight with depth using Eq. (4.2) and the
values in Table 4.1. Use effective unit weight in the saturated zone.
(f) Establish the overburden pressure distribution with depth by multiplying the unit
weight at a specific depth by the depth increment and adding the stress from the
previous increment. This causes the overburden pressure to compound with depth
while accounting for the variation of the unit weight.
(g) Plot the variation of positive earth pressure (i.e. σvKa) with depth. Use effective
stress in the saturated zone.
(h) Convert volumetric water content to the degree of saturation and plot the
variation of negative earth pressure (i.e. 2 aC K ) with depth.
(i) Plot the net AEP distribution as the difference between (g) and (h).
(j) Attain the regression curve for the AEP distribution.
45
(k) Using a tool such as Wolfram Alpha, integrate the regression curve (Figure 4.5)
from the ground surface to the depth where net AEP equals zero to determine A1
(i.e. resultant force in tension zone). Integrate from the depth where AEP = 0 to
the depth which provides A1 = A2 to find the critical height. This may involve
integrating beyond the depth of the GWT. In this case, the first part of A2 is
determined by integrating up to the depth of the GWT. The remaining area is
found by integrating the function for the saturated zone. The critical height is
taken as the depth from the ground surface to where A1 = A2.
Figure 4.5. Example of regression curve for the net AEP distribution with D = 0.6-m
Depth (m)
0.0 0.2 0.4 0.6 0.8 1.0
Net
act
ive
eart
h p
ress
ure
(kP
a)
-4
-2
0
2
4
6
y = 2.656x + 1.5743R² = 1
y = 42.553x4 - 81.394x3 + 57.779x2 - 5.5957x - 2.2089R² = 1
GW
T
Hcr
A1
A2
D
Hcr
46
Figure 4.6 to Figure 4.9 show the distributions of the positive, negative, and net AEP with
depth for D = 0.3, 0.6, 1.0, and 1.5-m, respectively. The absolute value of negative earth
pressure is plotted such that the magnitude of the positive and negative earth pressures can
be compared more easily. The negative sign in the net AEP distribution indicates the
tension zone, and the positive sign indicates the compression zone. When the lateral earth
pressure is zero, the stress-state is transitioning from tension to compression or vice versa.
When D = 1.0-m, the magnitude of negative earth pressure is not sufficient to match the
positive pressure. Also, in the case with D = 1.5-m, the net AEP is positive throughout the
entire soil profile. This indicates that an unsupported vertical trench fails immediately upon
initiating the excavation (i.e. Hcr = 0-m).
47
Figure 4.6. Positive, negative, and net AEP distribution (D = 0.3-m)
Pressure (kPa)
-3 -2 -1 0 1 2 3 4
Dep
th (
m)
0.0
0.1
0.2
0.3
0.4
0.5
Positive earth pressure (vKa)
Negative earth pressure ( )Net active earth pressure (vKa )
Zt
a- 2C Ka2C K
48
Figure 4.7. Positive, negative, and net AEP distribution (D = 0.6-m)
Pressure (kPa)
-4 -2 0 2 4 6
Dep
th (
m)
0.0
0.2
0.4
0.6
0.8
1.0
Positive earth pressure (vKa)
Negative earth pressure ( )Net active earth pressure (vKa )
Zt
a2C K
a- 2C K
49
Figure 4.8. Positive, negative, and net AEP distribution (D = 1.0-m)
Pressure (kPa)
-2 0 2 4 6
Dep
th (
m)
0.0
0.2
0.4
0.6
0.8
1.0
50
Figure 4.9. Positive, negative, and net AEP distribution (D = 1.5-m)
Pressure (kPa)
-3 0 3 6 9
Dep
th (
m)
0.0
0.3
0.6
0.9
1.2
1.5
Positive eath pressure (vKa)
Negative earth pressure ( ) Net active earth pressure (vKa )a- 2C K
a2C K
51
4.3 Estimating the Critical Height with BSM
In this section, the critical height of an unsupported vertical trench in sand was estimated
using Bishop’s simplified method with geotechnical modelling software, SLOPE/W.
Bishop (1955) assumed that the tangential interslice forces are equal and opposite (i.e. X1
= X2 in Figure 4.10) but horizontal forces are not. The FOS is calculated using Eq. (4.3).
Figure 4.10. Forces acting on the ith slice in Bishop’s simplified method of slices (Craig 2004)
sec1tan
tan tansin1
ii i i i i i
i i ii ii
FOS c W uW
FOS
(4.3)
where FOS = factor of safety, Wi = weight of ith slice, αi = inclination of ith slice base, and
β = width of ith slice
52
This method only satisfies moment equilibrium because interslice shear forces are ignored,
which implies rotational failure is assumed to happen about a circular slip surface (i.e.
negligible internal shearing occurs). This sort of mechanism was deemed plausible for a
vertical trench being excavated into sand with apparent cohesion. An iterative procedure is
required to solve for equilibrium. The initial guess for the FOS in SLOPE/W is one, in
which case the shear strength, and a new FOS are computed. The program continues
iterating until convergence occurs at a specified tolerance (i.e. 0.001 between values for
the FOS). The critical height was taken as the depth immediately preceding the depth that
showed FOS < 1 in the stability analysis.
In-situ analyses in SIGMA/W (Figure 4.11) are used for setting the initial PWP profile and
the ground stresses. In-situ analyses require Fixed-X displacement boundaries (i.e. hollow
red triangles) on the lateral extents, and Fixed-XY boundaries along the base of the domain.
The stresses calculated in SIGMA/W are not used in calculating the FOS for a Bishop
analysis in SLOPE/W; therefore, meshing and the material model are not important for this
application. However, an in-situ analysis was used as the parent to each SLOPE/W analysis
(Figure 4.12) to establish the PWP profile. This was done for convenience, such that the
GWT only needed to be changed in one analysis file rather than in each individual
excavation stage within SLOPE/W.
53
Figure 4.11. Establishing PWP profile with In-situ analysis in SIGMA/W
Figure 4.12. Analysis tree for Bishop’s simplified method with In-situ as the parent analysis
GWT
-0.46-0.44-0.42-0.40-0.38-0.36-0.34-0.32-0.30-0.28-0.26-0.24-0.22-0.20-0.18-0.16-0.14-0.12-0.10-0.08-0.06-0.04-0.02
Pore-Water Pressure
-3 - -2 kPa-2 - -1 kPa-1 - 0 kPa0 - 1 kPa1 - 2 kPa
Displacement boundary
(SIGMA/W)
(SLOPE/W)
54
In SIGMA/W, the initial PWP profile can be simulated by defining a GWT and a maximum
negative pressure head based on the assumption that PWP varies hydrostatically with
distance above and below the GWT (Figure 4.13). For instance, if the maximum negative
pressure head is lower than the height of the unsaturated soil layer (i.e. Hmax < Hunsat), the
negative PWP is constant up to ground surface beyond the maximum negative pressure
head. Conversely, if the maximum negative pressure head is greater than the height of the
unsaturated soil layer (i.e. Hmax > Hunsat) the negative PWP increases linearly up to ground
surface. Alternatively, a spatial function can be used to specify the pressure head at discrete
points. The pressure head is determined at intermediate points using linear interpolation.
In this study, an initial GWT was assigned to the problem domain to simulate hydrostatic
PWP distributions above and below the GWT. The maximum negative pressure head was
left as the default setting, 5-m.
u = (y - y )
A
B
C u = (y - y )
Hunsat = Height of unsaturated soil layer
yyw
u = Max. negative pressure
Hmax < Hunsat
Hmax
Hunsat
Hsat
Hmax = Max. negative pressure headHsat = Height of saturated soil layer
Hunsat>
a
b bw w
c cw w
Figure 4.13. Calculation of PWP based on the location of GWT in SIGMA/W
55
For soil with negative PWP (i.e. matric suction), total cohesion is computed with Eq. (4.4)
using effective cohesion and the SWCC. The residual volumetric water content was taken
as 5% of the volumetric water content at saturation for calculating total cohesion in
SLOPE/W. The Mohr-Coulomb failure criterion was used as the material model.
tanra w
s r
C c u u
(4.4)
In SLOPE/W, the shape and positions of trial slip surfaces can be defined by four options:
i) Entry and Exit, ii) Grid and Radius, iii) Block Specified, and iv) Fully Specified. Each
method processes every defined slip surface to find the minimum FOS. The ‘Entry and
Exit’ method was chosen to specify the trial slip surfaces in this chapter (Figure 4.14). The
entry was defined as a range as wide as the excavation depth with an entry point every 10-
mm, and the exit was specified as a point at the toe of the slope. Figure 4.15 to Figure 4.18
show the results for the stability analyses with D = 0.3, 0.6, 1.0, and 1.5-m, respectively.
The contours indicate PWP.
Figure 4.14. Schematic of the ‘Entry and Exit’ slip surface method in SLOPE/W
Radius points
Entry range
Exit point
56
Figure 4.15. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 0.3-m)
Figure 4.16. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 0.6-m)
-2
-1
1.02
-0.30-0.28-0.26-0.24-0.22-0.20-0.18-0.16-0.14-0.12-0.10-0.08-0.06-0.04-0.02
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
-5
-4
-3
-2
-1
1.01
-0.58-0.54-0.50-0.46-0.42-0.38-0.34-0.30-0.26-0.22-0.18-0.14-0.10-0.06-0.02
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
57
Figure 4.17. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 1.0-m)
Figure 4.18. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 1.5-m)
-9.5
-9
0.36
-0.09-0.08-0.07-0.06-0.05-0.04-0.03-0.02-0.01 Factor of Safety
≤ 0.35 - 0.450.45 - 0.550.55 - 0.650.65 - 0.75≥ 0.75
-14.5
-14
0.36
-0.08-0.07-0.06-0.05-0.04-0.03-0.02-0.01
Factor of Safety
≤ 0.35 - 0.450.45 - 0.550.55 - 0.650.65 - 0.75≥ 0.75
58
4.4 Comparison of Critical Heights from EREPT and BSM
Multiple analyses were conducted with the GWT at various elevations to establish a
relationship between the depth of the GWT and the critical height. Table 4.3 summarizes
the critical heights estimated using extended Rankine earth pressure theory and Bishop’s
simplified method. Based on the results in Table 4.3, the variation of the critical height
with respect to the depth of the GWT in Unimin 7030 sand is plotted in Figure 4.19.
Table 4.3. Variation of the critical height in Unimin 7030 sand
GWT (m)
Critical Height (m)
Extended Rankine Earth Pressure Theory Bishop's Simplified Method
0 0.00 0.00
0.1 0.11 0.12
0.2 0.20 0.23
0.3 0.30 0.35
0.4 0.39 0.46
0.5 0.48 0.57
0.6 0.57 0.65
0.7 0.64 0.70
0.8 0.68 0.71
0.9 0.00 0.04
1 0.00 0.00
1.1 0.00 0.00
1.2 0.00 0.00
1.3 0.00 0.00
1.4 0.00 0.00
1.5 0.00 0.00
59
The results from both methods show good agreement. The error ranged from 4%
(0.03/0.68-m) with D = 0.8-m, to 19% (0.09/0.48-m) with D = 0.5-m. Both methodologies
show that the critical height increases gradually as the depth of the GWT increases up to
0.8-m, and then drops to zero as the depth of the GWT is further increased. The minimum
slip surface depth was set at 0.01-m in SLOPE/W, therefore the FOS could not be generated
for the 0.01-m excavation stage. In which case, the critical height was taken as 0-m when
the 0.02-m excavation stage showed FOS < 1 (e.g. Figure 4.17 and Figure 4.18). This
behaviour can be explained by the variation of total cohesion in the sand with respect to
matric suction. The contribution of matric suction towards total cohesion increases with
increasing matric suction, and then starts decreasing as residual suction is approached (7.8-
kPa for Unimin 7030 sand; Vanapalli et al. 1996). In other words, total cohesion becomes
zero in the sand near the top of the trench when the depth of the GWT is relatively deep (>
0.8-m), and shear strength becomes fully dependant on frictional resistance. This indicates
that trench failure occurs at a shallow depth when the GWT is deep enough to create a
residual zone near the ground surface.
For depths of the GWT up to 0.7-m, the critical heights estimated using extended Rankine
earth pressure theory are slightly greater than the depths of the GWT. This may not be
realistic since the sand used in the present study does not have effective cohesion and an
excavation below the GWT can initiate a localized failure near the toe of the slope (i.e.
slough-in/cave-in, Figure 2.5(c)), which leads to a general failure. In other words, these
methods do not account for localized failures that may trigger general failures.
It is interesting to note that the results show close agreement even though the respective
methods are inherently different. Bishop’s simplified method only satisfies moment
60
equilibrium and assumes a circular slip surface, and the shear strength is computed at the
base of each slice along the slip surface. Extended Rankine earth pressure theory is purely
based on force equilibrium. The lateral pressure along the excavation face is computed and
the sum of the forces in the horizontal direction is set to zero to solve for the critical height.
The results are likely close because failure naturally occurs at a relatively shallow depth
and the soil mass is approximately the same regardless of the assumed failure surface.
Figure 4.19. Variation of the critical height with respect to the depth of the GWT using extended Rankine earth pressure theory and Bishop’s simplified method (Unimin 7030 sand)
Depth of GWT (m)
0.0 0.5 1.0 1.5
Hc
r (m
)
0.0
0.2
0.4
0.6
0.8
1.0
Extended Rankine TheoryBishop's Simplified Method
Critical Height = Depth of GWT
61
4.5 Summary and Conclusions
In this chapter, an attempt is made to estimate the critical height of an unsupported vertical
trench in an unsaturated sand using two methodologies; extended Rankine earth pressure
theory, and Bishop’s simplified method. The results showed that the critical height
increases with increasing depth of the GWT up to a point (0.8-m), and then decreases
significantly as the depth of the GWT is further increased. This is attributed to the fact that
total cohesion of an unsaturated sand becomes close to zero as matric suction exceeds the
residual suction value. The agreement between the results suggests that the critical height
of an unsupported vertical trench in unsaturated sand is strongly influenced by the variation
of shear strength with respect to matric suction.
62
CHAPTER 5
ESTIMATING THE CRITICAL HEIGHT OF UNSUPPORTED TRENCHES WITH DIFFERENT WALL SLOPES IN SAND
In Chapter 4, trench stability was estimated without considering the change in PWP due to
excavating. In practice, excavating a trench relieves stress in the soil within the proximity
of the excavation, which results in elastic rebound. This phenomenon leads to an increase
in void ratio and a temporary decrease in PWP. The shear strength of an unsaturated soil
can either increase or decrease depending on the initial matric suction distribution before
excavating. For example, if a trench is excavated into a soil where matric suction is close
to or within the residual zone, a further increase in matric suction can lead to a decrease in
shear strength and therefore the critical height. Conversely, if most soil above the GWT is
within the boundary effect or transition zone, a temporary decrease in PWP may
temporarily increase the shear strength of the soil. However, the probability of the trench
failing increases with time as negative PWP (i.e. matric suction) dissipates.
Coupled stress-PWP analyses (hereafter referred to as coupled analyses) are conducted in
this chapter to investigate the critical height of trenches considering the influence of PWP
redistribution caused by excavating. Four different wall slopes (1.5V:1H, 2V:1H, 3V:1H,
and 90°) and a vertical trench with the top 0.3-m sloped 1:1 (denoted as 90***, Figure 5.1)
were considered with multiple depths of the GWT (0, 0.3, 0.5, 0.7, 0.8, 0.9, 1.0, 1.2, 1.5,
and 2.0-m) in Unimin 7030 sand (Table 4.1). For comparison, the critical heights were also
estimated using the LEM (Morgenstern-Price method) for the same excavation scenarios
used in the coupled analyses.
63
Figure 5.1. Different types of trenches considered in this chapter
5.1 Estimating the Critical Height with the Finite Element Approach
5.1.1 Hydraulic Conductivity Function
The hydraulic conductivity of a soil is maximized when the soil is saturated. As a soil
desaturates, air gradually replaces the voids initially occupied by water and the tortuosity
of the flow path increases. This phenomenon increases the resistance to flow through the
voids. In other words, the hydraulic conductivity of a soil is a function of the degree of
saturation (or matric suction) and decreases with increasing matric suction. However,
measuring the hydraulic conductivity for various matric suction values is time-consuming
90°
90***
V
HV:H slope
64
and requires elaborate testing equipment. For this reason, the hydraulic conductivity
function proposed by Fredlund et al. (1994), Eq. (5.1), was used in this chapter to estimate
the variation of hydraulic conductivity with respect to soil suction. Figure 5.2 shows the
hydraulic conductivity function of Unimin 7030 sand.
1
i
i
i
i
yNy
yi j
unsat sat yNs y
yi
ee
ek k
ee
e
(5.1)
where
kunsat = the calculated conductivity for a specified water content or matric suction,
ksat = the measured conductivity for saturated condition,
y = a variable of integration representing the logarithm of negative PWP,
i = the interval between the range of j to N,
j = the least negative PWP to be described by the final function,
N = the maximum negative PWP to be described by the final function,
= the suction corresponding to the jth interval,
’ = the first derivative of Eq. (4.1)
65
Figure 5.2. Hydraulic conductivity function of Unimin 7030 sand
5.1.2 Analysis in SIGMA/W
The PWP distribution and the initial stresses (i.e. gravity body loads) first must be
established with an In-situ analysis prior to excavating (Figure 4.11). A hydrostatic PWP
distribution was created by setting a static GWT. The excavations were staged using a
coupled analysis. Deformations and stress distributions resulting from each excavation
stage were calculated using effective stress parameters and the elastic-plastic constitutive
model.
Figure 5.3 shows the boundary conditions required to perform a coupled analysis in
SIGMA/W. The left and right ends are restrained in the X-direction (i.e. hollow red
Matric Suction (kPa)
0.1 1.0 10.0
Hyd
rau
lic
Co
nd
uct
ivit
y, k
(m
/s)
10-10
10-9
10-8
10-7
10-6
10-5
10-4
66
triangles), and the base of the domain is restrained in both the X and Y directions. Total
head boundaries equal to the initial water table elevation were placed along the lateral
extents of the soil region (i.e. solid blue dots). This allows the GWT to fluctuate in response
to excavating while maintaining constant hydraulic head along the extents of the domain.
The mesh is created of 0.1 × 0.25-m elements in the immediate surroundings of the
excavation, and transitions to 1 × 1-m elements along the extents of the domain. A finite
element mesh pattern of ‘quads & triangles’ was used to provide a smooth transition
between areas of interest. The main reason for using different element sizes was to save on
computation time. Mesh sizes were determined based on a mesh-convergence study
conducted with different element lengths (1-m, 0.25-m, 0.1-m, and 0.05-m) and a 0.1-m
thickness along the excavated surface. The results showed that the critical height is not
affected by the mesh size when finer than 0.25-m. 4-point integration was used for the
quadrilateral elements, and 3-point integration was used for the triangular elements. A
linear interpolation model was used for calculating stresses and deformations at the nodes.
The use of secondary nodes was not necessary for this application.
The previous stage in the excavation was used as the parent analysis to the following, such
that the stress changes and deformations caused by the previous excavations were
compounded as the stages progressed. Excavations were simulated by deactivating regions
in 0.1-m increments. Analyses for each slope and GWT were conducted with 10-s and
1,000-s time steps between excavation stages to investigate the variation of PWP caused
by different excavation rates. Soil parameters used in the analyses are listed in Table 4.1.
67
Figure 5.3. Boundary conditions required for coupled analyses in SIGMA/W
5.1.3 Analysis in SLOPE/W
The FOS was calculated at targeted time steps based on the stress-deformation results from
SIGMA/W (i.e. SIGMA-Stress type analyses were conducted in SLOPE/W). An example
of the analysis tree for 10-s time steps is shown in Figure 5.4. Shear strength is computed
with either Eq. (3.4) or Eq. (3.14) depending on the location of the potential slip surface
and GWT. The mobilized and resisting shear forces are computed for each slice in finite
element analyses, therefore the FOS is computed for each slice. The global FOS is found
by summing all forces along the slip surface (Eq. (5.2)).
Total head boundary
Displacement boundary
Excavated area
68
r
m
SFOS
S
(5.2)
where Sr = total available shear resistance, and Sm = total mobilized shear force along the
base of the slice
Figure 5.4. Example of slope stability analysis tree used in the coupled analyses (10-s time steps between excavations)
time step
excavation stage
69
The ‘Entry and Exit’ slip surface method was used for generating potential slip surfaces.
The exit was specified as a point at the toe of the excavation, and the entry was defined as
a range as wide as the excavation depth with a possible entry point every 10-mm along the
ground surface (Figure 4.14). The critical height was defined as the excavation depth that
showed FOS = 1.0 (e.g. Figure 5.5), or the depth prior to the excavation stage that showed
FOS < 1.0 in the stability analysis. Contours in Figure 5.5 indicates PWP with depth.
Figure 5.5. Example of slope stability analysis using SIGMA-Stress method in SLOPE/W (1.5V:1H)
-8 -4
0
4
8
12
16
1.001
70
5.2 Estimating the Critical Height with the Limit Equilibrium Method
Stability analyses were also conducted with the LEM for the same scenarios used in the
coupled analyses to investigate the differences. Among various solutions to the LEM, the
method proposed by Morgenstern & Price (1965; hereafter referred to as M-P method) was
used in this chapter. Setting up the analysis domain in SLOPE/W for the M-P method is
the same as a Bishop analysis as described in Chapter 4 (Figure 4.11, Figure 4.12, and
Figure 4.14). The interslice forces are statically indeterminate with the LEM, therefore
various solutions exist based on the assumptions made to solve for equilibrium. In Bishop’s
simplified method, the interslice shear forces are assumed to be equal and opposite to solve
for equilibrium. However, the M-P method accounts for both interslice normal and shear
forces, assuming that the interslice shear force is a function of a scaling factor, an interslice
force function, and the interslice normal force, as shown in Eq. (5.3).
X f x E (5.3)
where X = interslice shear force per unit length, E = interslice normal force per unit length,
λ = scaling factor, and f(x) = specified interslice force function
The interslice force function describes how the magnitude of X/E varies across the failure
surface. The assumption regarding the interslice force function may lead to convergence
issues depending on the stress distribution and slope geometry (Ching & Fredlund 1983).
The half-sine interslice force function was used in this chapter, because it was chosen as
the default setting in SLOPE/W due to user experience and intuition (GEO-SLOPE 2015).
The half-sine function causes the slices on the ends of the slip surface to have the lowest
71
shear-to-normal stress ratio, and the slices in the middle have the highest (Figure 5.6). Once
f (x) is specified, λ is varied systematically to determine the value at which moment and
force equilibrium have the same FOS. The corresponding λ value is multiplied by the
specified f(x) to get the applied f(x), which provides the assumed shear-to-normal stress
ratio for each slice (Figure 5.6).
Figure 5.6. Example of half-sine function used in M-P analyses (λ = 0.25)
If λ equals zero, the M-P method becomes the same as Bishop’s simplified method (i.e.
internal shear forces are equal and opposite). Bishop’s simplified method is appropriate
when rotational failure is likely because no slippage between slices is required for a soil
mass to rotate along a circular slip surface. However, significant slippage and internal
shearing must occur for a soil mass to translate along a circular slip surface. Translational
failure becomes more likely in a cohesionless soil as the slope angle approaches the internal
friction angle. Therefore, the FOS can be unrealistic using Bishop’s simplified method for
Slice Number0 5 10 15 20 25 30
Inte
rslic
e F
orc
e F
un
ctio
n
0.0
0.2
0.4
0.6
0.8
1.0
Specified
Applied
72
sloped excavations in sand. This indicates that the M-P method is more appropriate to
analyze trench stability for the cases examined in this chapter. However, it should be noted
that the interslice and slip surface forces assumed in the LEM may not represent in-situ
stress conditions. Nevertheless, the global FOS from the M-P method is normally
acceptable and adequately addresses general failure (Krahn 2003).
Figure 5.7 shows the forces acting on a slice within an arbitrary slip surface and defines all
geometric parameters. Lateral pressure due to water in tension cracks (AR), external point
load (D), and seismic loads (kW), were not considered in the analyses and are therefore
omitted from the equations for calculating the FOS. In saturated soil, the FOS with respect
to moment and force equilibrium is calculated using Eq. (5.4) and Eq. (5.5), respectively.
The normal force is computed using Eq. (5.6).
Figure 5.7. Forces acting on a slice within an arbitrary slip surface (GEO-SLOPE 2015)
73
( ( ) tan )
m
c R N u RFOS
Wx Nf
(5.4)
( cos ( ) tan cos )
sinf
c N uFOS
N
(5.5)
sin sin tan( )
sin tancos
R L
c uW X X
FOSN
FOS
(5.6)
Where FOSm = factor of safety for moment equilibrium, FOSf = factor of safety for force
equilibrium, N = slice base normal force per unit length (FOS = FOSm or FOSf)
Extending Eq. (5.4) and (5.5), SLOPE/W computes the FOS in unsaturated soil for
moment and force equilibrium as shown in Eq. (5.7) and (5.8), respectively (Fredlund &
Krahn 1977, Fredlund et al. 1981). The normal force, N, is computed as shown in Eq. (5.9).
tan tan1 tan
tan tan
b b
w a
m
c R N u u R
FOSWx Nf
(5.7)
tan tancos 1 tan cos
tan tan
sin
b b
w a
f
c N u u
FOSN
(5.8)
sin sin (tan tan ) sin tan( )
sin tancos
b ba w
R L
c u uW X X
FOSN
FOS
(5.9)
74
5.3 Comparison of Limit Equilibrium and Finite Element Approaches
The FOS of a trench may be relatively high immediately after excavating but may decrease
over time as the GWT and PWP return to equilibrium. This sort of scenario may occur in
practice if an excavation is made rapidly to a desired depth and left open for some period.
Figure 5.8 shows the variation of deformation, PWP, and FOS with time for the case of a
1.5V:1H sloped excavation staged in 0.1-m increments up to 1.3-m at 10-s time steps, with
the GWT initially set at 0.7-m. The time step for the 1.3-m stage was increased to 1,000-s
(Figure 5.4) to allow the GWT to rebound. The black arrows represent hydraulic velocity
vectors, and the magnitude decreases with time as PWP approaches equilibrium. The
sequence in Figure 5.8 shows how an excavation may appear stable but fail shortly
thereafter (i.e. 750-s in this example). As discussed previously, removing soil from the
ground relieves confining pressures which results in expansion of the soil adjacent to the
excavated surface. Figure 5.8 clearly shows that the deformations along the excavation face
gradually increase over time. This phenomenon contributes to the decrease in FOS with
time as shown in Figure 5.9.
75
10
-6
2
-2
2
6
14
Dep
th (
m)
-3.0-2.8-2.6-2.4-2.2-2.0-1.8-1.6-1.4-1.2-1.0-0.8-0.6-0.4-0.20.0
4
8
12
16
-4
Dep
th (
m)
-3.0-2.8-2.6-2.4-2.2-2.0-1.8-1.6-1.4-1.2-1.0-0.8-0.6-0.4-0.20.0
-6
2
6
10
14
18
-2
De
pth
(m
)
-3.0-2.8-2.6-2.4-2.2-2.0-1.8-1.6-1.4-1.2-1.0-0.8-0.6-0.4-0.20.0
4
8
12
16
1.30
Dep
th (
m)
-3.0
-2.7
-2.4
-2.1
-1.8
-1.5
-1.2
-0.9
-0.6
-0.3
0.0
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
4
8
12
16
20
1.12
Dep
th (
m)
-3.0
-2.7
-2.4
-2.1
-1.8
-1.5
-1.2
-0.9
-0.6
-0.3
0.0
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
4
8
12
16
1.32
Dep
th (
m)
-3.0
-2.7
-2.4
-2.1
-1.8
-1.5
-1.2
-0.9
-0.6
-0.3
0.0
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
(c) 500s
(b) 250s
(a) 10s
76
Figure 5.8. Variation of deformation, PWP, and FOS with time for (a) 10-s, (b) 250-s, (c) 500-s, and (d) 750-s after 1.3-m excavation stage with initial D = 0.7-m
Figure 5.9. FOS vs. Time for 1.3-m excavation stage (1.5V:1H) with initial D = 0.7-m
Time (s)
0 200 400 600 800 1000 1200
FO
S
0.9
1.0
1.1
1.2
1.3
1.4
1.5
excavation
-6
2
6
10
14
18
-2
Dep
th (
m)
-3.0-2.8-2.6-2.4-2.2-2.0-1.8-1.6-1.4-1.2-1.0-0.8-0.6-0.4-0.20.0
4
8
12
16
20
1.00
De
pth
(m
)
-3.0
-2.7
-2.4
-2.1
-1.8
-1.5
-1.2
-0.9
-0.6
-0.3
0.0
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
(d)
77
Negligible changes in matric suction are observed near the ground surface throughout the
time steps used in the analysis. This implies that the slope failure is governed by the change
in matric suction distribution between the initial GWT and the toe of the slope. The FOS
remains relatively constant during the first 250-s because the length of slip surface below
the GWT is relatively unchanged, even though the GWT increases with time. However,
the FOS drops rapidly with time once the length of slip surface below the GWT exceeds
what can be considered the critical length of the slip surface under saturated condition.
In geotechnical engineering practice, trenches may be excavated at a relatively fast rate
between stages without allowing the GWT to rebound to its original level. However, once
a trench is excavated up to the targeted depth, PWP will gradually approach equilibrium.
This may occur while field workers are completing their jobs in the trench. As shown in
Figure 5.9, a trench may appear stable initially but can fail over time as the GWT rebounds.
It is interesting to note that failure can occur even before PWP reaches equilibrium. Figure
5.8(d) shows FOS = 1 after 750-s without equilibrium being reached, as evidenced by the
hydraulic velocity vectors.
To further investigate how the redistribution of PWP caused by excavating affects the
critical height of unsupported trenches, slope stability analyses were carried out for two
different time steps; namely, 10-s and 1000-s. These time steps were chosen to simulate
fast and slow excavation rates, respectively. Figure 5.10 and Figure 5.11 show the variation
of the critical height with respect to the depth of the GWT for five different excavation
scenarios with 10-s and 1,000-s time steps, respectively. Figure 5.12 shows the variation
of the critical heights estimated using the M-P method for the same scenarios used in the
coupled analyses. The critical height increases as trench slope is reduced regardless of the
78
level of the GWT. This indicates that sloping trenches is an effective way to increase the
critical height. For 10-s time steps, the critical height of the 1.5V:1H slope was estimated
to be 2.6 times higher compared to the vertical trench for some cases. The critical height
of the vertical trench increases as the depth of the GWT increases and then declines sharply
once the GWT is deeper than 1-m.
The critical heights estimated from the three methods are plotted for the 90°, 3V:1H, 2V:1H,
and 1.5V:1H excavation scenarios in Figure 5.13 to Figure 5.16, respectively (90*** is
explained later).
Figure 5.10. Variation of the critical height for different sloping angles in Unimin 7030 sand (10-s time steps, coupled analysis)
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.59090*** 3V:1H2V:1H1.5V:1H
79
Figure 5.11. Variation of the critical height for different sloping angles in Unimin 7030 sand (1,000-s time steps, coupled analysis)
Figure 5.12. Variation of the critical height for different sloping angles in Unimin 7030 sand (M-P method)
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.59090*** 3V:1H2V:1H1.5V:1H
GWT (m)0.0 0.5 1.0 1.5 2.0
Hcr
(m
)
0.0
0.5
1.0
1.5
2.0
2.5
9090*** 3V:1H2V:1H1.5V:1H
80
Figure 5.13. Variation of the critical height for 90° slope in Unimin 7030 sand
Figure 5.14. Variation of the critical height for 3V:1H slope in Unimin 7030 sand
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.510s1,000sM-P
GWT (m)0.0 0.5 1.0 1.5 2.0
Hcr
(m
)
0.0
0.5
1.0
1.5
2.0
2.510s1,000sM-P
81
Figure 5.15. Variation of the critical height for 2V:1H slope in Unimin 7030 sand
Figure 5.16. Variation of the critical height for 1.5V:1H slope in Unimin 7030 sand
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.510s1,000sM-P
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.510s1,000sM-P
82
As expected, the critical heights estimated using the M-P method showed the lowest values.
The critical height decreases as the period between excavation stages increases (i.e. 10-s to
1,000-s time steps). This is because 10-s time step excavations force the GWT to continue
dropping without giving enough time to rebound a significant amount.
When a soil mass is removed from the ground, the soil within the proximity of the
excavation experiences stress relief ranging from K0z to z in the horizontal and vertical
directions, respectively (where z is the vertical stress at a depth z, and K0 is the coefficient
of earth pressure at-rest). This indicates that the stress relief in the soil near the ground
surface is minimal (or close to zero), and PWP changes are negligible. Therefore, as the
slope of a trench decreases, more stress relief occurs along the face of the slopes and the
change in PWP increases accordingly. For this reason, the largest difference in the critical
heights between the coupled analysis (10-s) and the M-P method is observed for the
1.5V:1H slope. For coupled analyses with 1,000-s time steps, most negative excess PWP
dissipates and the critical heights show small discrepancies with the M-P method when the
GWT is less than 1-m. When the GWT is deeper than 1-m, the critical heights estimated
with the M-P method are zero, but the coupled analyses show minimum values of 0.3-m.
Stability analyses based on limit equilibrium conditions are normally more conservative
compared with that of finite element stress-based analyses. However, there are advantages
to using finite element stress-based stability analyses such as i) displacement compatibility
is satisfied and ii) the ground stresses are much closer to reality (Krahn 2003). The main
reason for this difference is related to the concentration of normal and shear stress in the
toe area of a slope as shown in Figure 5.17 and Figure 5.18, respectively. This phenomenon
83
may increase frictional resistance and in turn increase the local FOS in the slices near the
toe of the slope. For this reason, a sloped trench can be can be excavated past the residual
zone near the ground surface even when the GWT is relatively deep (i.e. deeper than 1-m).
The critical height increases for the vertical excavation as the depth of the GWT increases
up to 0.8-m, and then declines sharply thereafter. As explained previously, a deep GWT
creates a residual suction zone near the ground surface which results in a complete loss of
total cohesion (Figure 5.19). To further investigate this behavior, additional analyses were
conducted for a vertical trench with the first 0.3-m of excavation sloped 1:1 (denoted as
90***) as shown in Figure 5.20. The results from the three methods for the 90*** case are
plotted in Figure 5.21. For the M-P method, the maximum attainable critical height in a
vertical trench was increased by 0.1-m and was sustained for an additional 0.1-m increase
in the depth of the GWT by sloping the soil in the residual zone. For the coupled analyses,
the 90*** case provides higher critical heights with the GWT between 0.5 and 1.5-m. This
indicates that benching or sloping the soil in the residual zone can effectively increase the
critical height. Furthermore, sloping the residual zone may be more effective in some cases
than sloping the entire trench wall (Figure 5.10, Figure 5.11, and Figure 5.12).
84
Figure 5.17. Distribution of total normal stress during 0.3-m excavation stage for 1.5V:1H slope with initial D = 1.2-m (10-s time steps)
Figure 5.18. Distribution of shear stress during 0.3-m excavation stage for 1.5V:1H slope with initial D = 1.2-m (10-s time steps)
Dep
th (
m)
-1.2
-0.9
-0.6
-0.3
0.0
Max. Total Stress
≤ 0 - 2 kPa2 - 4 kPa4 - 6 kPa6 - 8 kPa8 - 10 kPa10 - 12 kPa12 - 14 kPa14 - 16 kPa16 - 18 kPa18 - 20 kPa≥ 20 kPa
Stress concentration
Dep
th (
m)
-1.2
-0.9
-0.6
-0.3
0.0
Max. Shear Stress
≤ 0 - 0.5 kPa0.5 - 1 kPa1 - 1.5 kPa1.5 - 2 kPa2 - 2.5 kPa2.5 - 3 kPa3 - 3.5 kPa3.5 - 4 kPa4 - 4.5 kPa4.5 - 5 kPa≥ 5 kPa
Stress concentration
85
Figure 5.19. Vertical trench with D = 0.9-m (M-P method)
Figure 5.20. Partially sloped (1:1) vertical trench with D = 0.9-m (M-P method)
-6
-4
-2
0.55
Dep
th (
m)
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
-0.0Factor of Safety
0.55 - 0.650.65 - 0.750.75 - 0.850.85 - 0.950.95 - 1.051.05 - 1.151.15 - 1.251.25 - 1.351.35 - 1.45≥ 1.45
-2
-6
2
6
0.90
Dep
th (
m)
-0.9
-0.7
-0.5
-0.3
-0.1Factor of Safety
0.90 - 1.001.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.60≥ 1.60
86
Figure 5.21. Variation of the critical height for 90*** slope in Unimin 7030 sand
5.4 Summary and Conclusions
M-P and coupled analysis methods were used to investigate the influence of various slopes
and matric suction distributions on the critical height of unsupported trenches in an
unsaturated sand. The coupled analyses show that a trench may appear stable initially and
result in failure as PWP returns to equilibrium conditions. Thus, it is more conservative to
assume a hydrostatic PWP distribution when estimating the critical height. Overall, sloping
the trench greatly increases the critical height. This lends credibility to Canadian provincial
standards (Ontario 2017), as they state a 1:1 slope is sufficient for most excavations where
a vertical trench cannot be excavated to a significant depth. The results also show that the
critical height can be almost doubled by flattening the slopes compared to a vertical
excavation. The main advantage of flattening the slope is that a greater critical height can
GWT (m)0.0 0.5 1.0 1.5 2.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.510s1,000sM-P
87
be achieved even when residual suction is reached near the ground surface. The critical
height was increased as the slope angle was reduced; however, it may not always be
possible to provide a gentle slope to the excavation in geotechnical engineering practice
due to limited space. In this case, a combination of sloping and benching can be used
effectively to achieve targeted critical heights, especially when the soil surface is within
the range of residual suction.
88
CHAPTER 6
ESTIMATING THE CRITICAL HEIGHT OF UNSUPPORTED VERTICAL TRENCHES SUBJECTED TO SURCHARGE PRESSURE
In this chapter, the critical height of unsupported vertical trenches in an unsaturated fine-
grained soil is investigated considering practical scenarios. The trenches were assumed to
be excavated into Indian Head till (hereafter referred to as IHT).
In well-developed cities, there are circumstances where trenches must be excavated
adjacent to an existing superstructure to install or remediate municipal infrastructure such
as storm drains or utility lines. In these cases, the critical height of unsupported vertical
trenches is affected by the stress beneath an existing shallow foundation (hereafter referred
to as foundation stress). Hence, analyses were conducted for multiple foundation stresses
(10, 20, 30, and 50-kPa) located at various distances from the edge of an excavation (1, 2,
3, and 5-m) to simulate this practical scenario. Each scenario was modelled with the GWT
at a depth of 0, 1, 2, 3, and 5-m below the shallow foundation (referred to as depth, D).
6.1 Soil Properties
Basic soil properties of IHT are shown in Table 6.1. The grain size distribution curve and
the SWCC of IHT are shown in Figure 6.1 and Figure 6.2, respectively. The SWCC was
best-fitted using Eq. (4.1) and the parameters, a, m, n are shown in the figure. The shear
strength parameters, c’ and ’ determined by Vanapalli et al. (1997) and Oh &Vanapalli
(2010) were used in the analyses.
89
Table 6.1. Basic soil properties of IHT (Vanapalli et al. 1997, Oh & Vanapalli 2010)
Properties Value
Plasticity Index, Ip (%) 15.5
Saturated unit weight, sat (kN/m3) 20.7
Void ratio, e 0.55
Specific gravity, Gs 2.72
Effective cohesion, c’ (kPa) 5
Effective internal friction angle, ’ (˚) 23.1
Figure 6.1. Grain size distribution curve of IHT (Oh &Vanapalli 2010)
Particle size (mm)
0.001 0.01 0.1 1 10
Per
cen
t fi
ner
(%
)
0
10
20
30
40
50
60
70
80
90
100
90
Figure 6.2. SWCC of IHT (Oh & Vanapalli 2018)
6.2 Estimating the Critical Height without Foundation Stress
In this section, critical heights are estimated using extended Rankine earth pressure theory
and Bishop’s simplified method (see sections 4.2 and Error! Reference source not found.
for more details) without considering foundation stress. This scenario represents the case
where an arbitrary foundation stress is sufficiently far away from the trench such that the
critical height of an unsupported vertical trench is not affected by the foundation.
6.2.1 Estimating the Critical Height with EREPT
The positive, negative, and net AEP distributions for cases with D = 0, 1, 3, and 5-m are
shown in Figure 6.3 to Figure 6.6, respectively.
Suction (kPa)
0.1 1 10 100 1000
Deg
ree
of
satu
rati
on
0.0
0.2
0.4
0.6
0.8
1.0
Fredlund and Xing (1994)Measured (Pressure plate)
1
ln
11.687
0.58066
1.4073
m
e nS
e a
a
m
n
91
Figure 6.3. Positive, negative, and net AEP distribution (D = 0-m)
Pressure (kPa)
-10 -5 0 5 10 15
Dep
th (
m)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Positive earth pressure (vKa)
Negative earth pressure (2C Ka)
Net earth pressure (vKa - 2C Ka)
zt
92
Figure 6.4. Positive, negative, and net AEP distribution (D = 1-m)
Pressure (kPa)
-12 -8 -4 0 4 8 12 16
Dep
th (
m)
0.0
0.5
1.0
1.5
2.0
zt
Positive earth pressure (vKa)
Negative earth pressure (2C Ka)
Net earth pressure (vKa - 2C Ka)
93
Figure 6.5. Positive, negative, and net AEP distribution (D = 3-m)
Pressure (kPa)
-20 -10 0 10 20 30
Dep
th (
m)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
zt
Positive earth pressure (vKa)
Negative earth pressure (2C Ka)
Net earth pressure (vKa - 2C Ka)
94
Figure 6.6. Positive, negative, and net AEP distribution (D = 5-m)
Pressure (kPa)
-20 -10 0 10 20 30 40 50
Dep
th (
m)
0
1
2
3
4
5
Positive earth pressure (vKa)
Negative earth pressure (2C Ka)
Net earth pressure (vKa - 2C Ka)
zt
95
6.2.2 Estimating the Critical Height with BSM
Figure 6.7 to Figure 6.10 show the slope stability analyses using Bishop’s simplified
method with D = 0, 1, 3, and 5-m, respectively.
Figure 6.7. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 0-m)
Figure 6.8. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 1-m)
5
10
15
20
25
1.01 2
.6 m
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
0 0 0
5 5
10
-5
1.00
1.8
m
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
96
Figure 6.9. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 3-m)
Figure 6.10. Slope stability analysis in SLOPE/W using Bishop’s simplified method (D = 5-m)
-25
-20
-15
-10
-5
1.04
2.5
m
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
-45
-40
-35
-30
-25
-20
-15
1.01
3.3
m
Factor of Safety
≤ 1.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.801.80 - 1.90≥ 1.90
97
6.2.3 Comparison of Critical Heights from EREPT and BSM
The critical heights obtained from extended Rankine earth pressure theory and Bishop’s
simplified method are summarized in Table 6.2 for various levels of the GWT. The results
are plotted in Figure 6.11 for better comparison.
Table 6.2. Variation of the critical height in IHT
Depth of GWT (m)
Critical Height (m)
Extended Rankine Earth Pressure Theory Bishop's Simplified Method
0 2.6 2.8
0.3 2.2 2.3
0.5 2.0 2.1
0.7 1.9 2.0
1 1.8 1.9
1.2 1.8 2.0
1.5 1.9 2.0
2 2.0 2.2
2.5 2.3 2.4
3 2.5 2.6
4 3.0 2.9
5 3.3 3.1
98
Figure 6.11. Variation of the critical height with respect to the depth of the GWT using extended Rankine earth pressure theory and Bishop’s simplified method (IHT)
The results show that there is good agreement between the critical heights obtained from
extended Rankine earth pressure theory and Bishop’s simplified method. This suggests that
extended Rankine earth pressure theory is a simple and efficient way to estimate the critical
height of unsupported vertical trenches in both coarse and fine-grained unsaturated soils.
It is interesting to note that the critical height does not continuously increase with
increasing depth of the GWT, but instead decreases gradually to a minimum value when D
= 1-m, and then starts increasing.
Depth of GWT (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Extended Rankine Theory
Bishop's Simplified Method
Critical Height = Depth of GWT
99
In a cohesive soil, net AEP is the sum of positive and negative earth pressure that are caused
by effective weight and total cohesion of the soil, respectively. The SWCC of a cohesive
soil is distributed over a large range of suction and the rate of change in the total cohesion
to that of soil suction is less compared to sandy soils. Hence, when the GWT is at a shallow
depth (i.e. less than 1-m), the contribution of total cohesion towards negative earth pressure
is less than that of effective weight of soil towards positive earth pressure. This becomes
opposite as the GWT is further decreased, which leads to an increase in the critical height.
In the case where the GWT is at the surface, total cohesion is minimized (= effective
cohesion); however, the unit weight of soil is minimized at the same time (i.e. effective
unit weight). For this reason, the critical height with D = 0-m is greater than that with the
GWT at a depth less than 3.5-m in this soil.
6.3 Estimating the Critical Height with Foundation Stress
The variation of the critical height of an unsupported vertical trench subjected to foundation
stress was estimated with the M-P method (details in section 5.2) to simulate excavations
nearby existing foundations. It was assumed that all soil was removed up to the wall of the
structure before proceeding to excavate below the foundation depth regardless of the
foundation stress. The scenario was modelled as shown in Figure 6.12.
100
Figure 6.12. SLOPE/W model used to consider foundation stress
The critical heights were estimated considering three variables; i) level of the GWT (0, 1,
2, 3 and 5-m), ii) foundation stress (10, 20, 30 and 50 kPa), and iii) distance between the
excavation and the foundation stress (0, 1, 2, and 3-m). The entry range of potential slip
surfaces was defined as double the depth of excavation, spanning from the edge of
excavation towards the foundation with an entry point every 10-cm. The exit was specified
as a point at the toe of the slope. An example of a stability analysis is shown in Figure 6.13.
Figure 6.14 to Figure 6.18 show the variation of the critical height with respect to the
distance from the edge of excavation to the foundation for different GWTs.
Distance (m)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Dep
th (
m)
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
Foundation stress
Distance between trench and foundation
101
Figure 6.13. Stability analysis considering 50-kPa surcharge at a distance of 3-m (D = 5-m)
Figure 6.14. Critical height vs. distance of foundation stress from excavation (D = 0-m)
0.98
Distance (m)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Dep
th (
m)
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
Factor of Safety
≤ 0.90 - 1.001.00 - 1.101.10 - 1.201.20 - 1.301.30 - 1.401.40 - 1.501.50 - 1.601.60 - 1.701.70 - 1.80≥ 1.80
Distance (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
10 kPa20 kPa30 kPa50 kPa
102
Figure 6.15. Critical height vs. distance of foundation stress from excavation (D = 1-m)
Figure 6.16. Critical height vs. distance of foundation stress from excavation (D = 2-m)
Distance (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
10 kPa20 kPa30 kPa50 kPa
Distance (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
10 kPa20 kPa30 kPa50 kPa
103
Figure 6.17. Critical height vs. distance of foundation stress from excavation (D = 3-m)
Figure 6.18. Critical height vs. distance of foundation stress from excavation (D = 5-m)
Distance (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
10 kPa20 kPa30 kPa50 kPa
Distance (m)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Hc
r (m
)
0
1
2
3
4
10 kPa20 kPa30 kPa50 kPa
104
The estimated critical heights decrease with increasing foundation stress. The distance at
which the curves converge in Figure 6.14 to Figure 6.18 shall be referred to as the
‘significant distance’. This is the distance at which the critical height is not affected by the
foundation stress. In other words, failure occurs due to the self-weight of the soil
overcoming the resistance from total cohesion, rather than being influenced by the
foundation stress when the significant distance is reached. For this reason, the critical
heights obtained at the significant distance show good agreement (maximum 8%
discrepancy) compared to those found without any applied foundation stress (Figure 6.11).
This discrepancy exists because different LEMs were used to estimate the critical heights
in section 6.2 and 6.3 (i.e. Bishop’s simplified method and M-P method). The significant
distance is approximately 3-m, 2-m, and 3-m, when D = 0-m, 1-m, and 5-m, respectively.
The shallowest critical height occurs when D = 1-m without any applied foundation stress,
therefore the significant distance is also lowest when D = 1-m. This is because the
significant distance is a function of the desired excavation depth in addition to applied
surcharge pressure (i.e. foundation stress).
The critical height vs. depth of the GWT is plotted in Figure 6.19 to Figure 6.22 for
different foundation stresses that are 0, 1, 2, and 3-m away from the excavation,
respectively. Except for when the distance is 0-m, the critical height is at a minimum for D
= 1-m, which is consistent with the case of no foundation stress. Figure 6.22 clearly shows
that the significant distance is no greater than 3-m in IHT for the foundation stresses and
levels of the GWT used in this chapter, as indicated by the overlapping surcharge curves.
105
Figure 6.19. Critical height vs. depth of the GWT with distance of foundation stress from excavation = 0-m
Figure 6.20. Critical height vs. depth of the GWT with distance of foundation stress from excavation = 1-m
Depth of GWT (m)
0 1 2 3 4 5
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
10 kPa20 kPa30 kPa50 kPa
Depth of GWT (m)
0 1 2 3 4 5
Hcr
(m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
10 kPa20 kPa30 kPa50 kPa
106
Figure 6.21. Critical height vs. depth of the GWT with distance of foundation stress from excavation = 2-m
Figure 6.22. Critical height vs. depth of the GWT with distance of foundation stress from excavation = 3-m
Depth of GWT (m)
0 1 2 3 4 5
Hc
r (m
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
10 kPa20 kPa30 kPa50 kPa
Depth of GWT (m)
0 1 2 3 4 5
Hc
r (m
)
0
1
2
3
4
10 kPa20 kPa30 kPa50 kPa
107
6.4 Summary and Conclusions
The variation of the critical height of an unsupported vertical trench was investigated with
respect to the proximity and magnitude of adjacent foundation stresses. The critical height
is less dependant on the magnitude of the foundation stress as the distance from the edge
of the excavation to the foundation stress is increased. The significant distance was
estimated to be approximately 3-m regardless of the level of the GWT used in this chapter.
The results show that in cases where a trench is excavated near existing superstructures, a
combination of the foundation stress, distance from the excavation to the foundation, and
level of the GWT (i.e. matric suction distribution) should be considered to reliably estimate
the critical height.
108
CHAPTER 7
GENERAL CONCLUSIONS
Being able to assess the critical height helps facilitate more efficient production in trench-
related work. Canadian provinces enforce strict regulations for excavations that workers
are required to enter (e.g. Regulation 181(1) of the Occupational Health & Safety Act
(1991)). In the case where workers are not required to enter the excavation, the regulations
state that contractors can use shoring at their own discretion. For example, machinery alone
may be capable of performing the task in the trench and no workers are subjected to a
collapse; however, a collapse can still damage the equipment and/or disrupt the work. Thus,
knowing the critical height for a jobsite allows the contractor to make informed decisions
with respect to supporting the excavation walls. Although this study provides framework
for estimating the critical height of unsupported trenches, it is recommended to use shoring
if soil deformations in the proximity of a trench (which can damage existing foundations
or utilities) is a concern.
An attempt was made to investigate the influence of the matric suction distribution on the
critical height of unsupported trenches in unsaturated soil. Extended Rankine earth pressure
theory and Bishop’s simplified method were used to independently estimate the variation
of the critical height of an unsupported vertical trench with respect to the depth of the GWT
in Unimin 7030 sand (Chapter 4), and Indian Head till (Chapter 6). The soils have
inherently different shear strength properties (cohesionless and cohesive soils), yet the
critical heights estimated using both methods showed good agreement for both soils. This
109
suggests that extended Rankine earth pressure theory and Bishop’s simplified method are
reliable methods for estimating the critical height of unsupported vertical trenches in
unsaturated soils where it is necessary to extend the effective stress approach.
Coupled stress analyses that account for changes in PWP over time were used to investigate
the critical height of unsupported excavations with various wall slopes and matric suction
distributions in Unimin 7030 sand. The results showed that the FOS is relatively high
immediately after the excavation, and then decreases over time as the PWP approaches
equilibrium conditions. The critical height increases with increasing depth of the GWT up
to a point and then decreases rapidly as the depth of the GWT is further increased,
regardless of the wall slope. This is due to the decrease in the contribution of matric suction
towards shear strength as matric suction of a soil near the ground surface reaches the
residual condition. The critical height was increased as the slope angle was reduced;
however, it may not be always possible to provide a gentle slope to the excavation in
geotechnical engineering practice due to limited work space in the field. In this case, a
combination of sloping and benching can be used effectively to achieve targeted critical
heights, especially when matric suction nearby the soil surface is in the range of residual
suction.
The Morgenstern-Price method was used in SLOPE/W to examine the influence of
foundation stress, the level of the GWT, and the distance between a trench and a foundation.
All soil should be removed up to the wall of the structure before proceeding with an
excavation below the foundation depth. The significant distance (i.e. the distance that the
critical height is not affected by the foundation stress) was estimated to be approximately
110
3-m for the foundation stresses and levels of the GWT used in this study. The significant
distance is a function of excavation depth and the applied surcharge pressure.
Most Canadian provinces impose a maximum allowable depth of 1.2-m for unsupported
vertical trenches, regardless of soil type and field conditions. The results from this study
suggest that imposing a universal safe height is flawed, and the safe height should be
estimated considering soil type, field conditions, and practical scenarios. In Unimin 7030
sand, only a 0.9-m critical height is the best-case scenario for an unsupported vertical
trench, which is less than the safe height recommended by any Canadian province.
Conversely, the worst-case scenario for the critical height in Indian Head till (i.e. 1.8-m) is
greater than the recommended safe height in all Canadian provinces.
In the case where the critical height is much less than what Canadian provinces suggest,
the critical height can be further increased by benching or sloping the trench walls,
especially in the residual zone. Ontario standards (2017) suggest that sloping the walls 1:1
from the base of excavation where it is not possible to excavate vertically to any significant
depth, and the findings from the current study reinforce this suggestion.
In the long-term, the effective cohesion of soil can be lost due to excessive strain, and
apparent cohesion may be lost due to excessive rainfall or evaporation. However, it is
unrealistic to omit these components of strength in the short-term design of unsupported
trenches.
111
7.1 Recommendations for Future Research
The long-term objective of this study is to provide field engineers and contractors with
guidelines for estimating the critical height of unsupported trenches with given field
conditions (i.e. soil type, matric suction distributions) and circumstances (e.g. surcharge
pressure). This research may facilitate more efficient trench related work by assisting
contractors in determining the depth at which shoring or a trench box is required to perform
any necessary tasks.
The findings from this study are based on analytical and numerical analyses, hence it is
necessary to conduct a series of laboratory and/or field tests to validate the findings. There
are also various practical scenarios that should be considered before estimating the critical
height of unsupported trenches for a jobsite by extending the findings from this study. For
example, if a trench must extend deeper than the GWT, de-watering the trench by pumping
induces seepage through the sidewalls of the trench. Seepage forces can decrease the
effective stress at the toe of the slope, and in turn lower shear strength. Another common
scenario that needs to be considered is heavy equipment operating near the top of a slope.
This induces vibration and affects the PWP and matric suction distribution within the soil.
Thus, the influence of seepage forces and dynamic loading should be studied to improve
the reliability of the methodologies for estimating the critical height of an unsupported
vertical trench.
112
REFERENCES
Alberta - Occupational Health & Safety Division. 2009. Occupational Health & Safety
Code - Explanation Guide.
Bishop, A.W. 1955. The Use of the Slip Circle in the Stability Analysis of Slopes.
Geotechnique, 5(1): 7–17.
Bishop, A.W. 1959. The Principle of Effective Stress. Teknisk ukeblad, 106(39): 859–863.
Bishop, A.W., and Blight, G.E. 1963. Some Aspects of Effective Stress in Saturated and
Partly Saturated Soils. Géotechnique, 13(3): 177–197.
Bowles, J.E. 2001. Foundation Analysis and Design. In 5th edition. McGraw-Hill.
Ching, R.K.H., and Fredlund, D.G. 1983. Some difficulties associated with the limit
equilibrium method of slices. Canadian Geotechnical Journal, 20(4): 661–672.
Craig, R.F. 2004. Craig’s Soil Mechanics. In 7th edition. Spon Press, New York.
Dai, B.B., Yang, J., and Zhou, C.Y. 2016. Observed Effects of Interparticle Friction and
Particle Size on Shear Behavior of Granular Materials. International Journal of
Geomechanics, 16(1): 1–11.
De Vita, P., Angrisani, A.C., and Di Clemente, E. 2008. Engineering geological properties
of the Phlegraean pozzolan soil (Campania region, Italy) and effect of the suction
on the stability of cut slopes. Italian Journal of Engineering Geology and
Environment, 2: 5–22.
Donald, I.B. 1957. Effective stresses in unsaturated non-cohesive soils with controlled
negative pore pressure. University of Melbourne, Melbourne, Australia.
113
Eivemark, M., and Hall, B. 2000. Geotechnical Considerations in Trench Design for
Lifelines. Jacques Whitford & Associates Ltd.
Escario, V., and Saez, J. 1987. Shear strength of partly saturated soils versus suction. In
International Conference on Expansive Soils. pp. 141–143.
Fredlund, D.G., and Krahn, J. 1977. Comparison of slope stability methods of analysis.
Canadian Geotechnical Journal, 14(3): 429–439.
Fredlund, D.G., Krahn, J., and Pufahl, D. 1981. The relationship between limit equilibrium
slope stability methods. In Tenth International Conference on Soil Mechanics and
Foundation Engineering. Stockholm. pp. 409–416.
Fredlund, D.G., and Morgenstern, N.R. 1977. Stress State Variables for Unsaturated Soils.
Journal of the Geotechnical Engineering Division, 103(5): 447–466.
Fredlund, D.G., Morgenstern, N.R., and Widger, R.A. 1978. The shear strength of
unsaturated soils. Canadian Geotechnical Journal, 15(3): 313–321.
Fredlund, D.G., and Rahardjo, H. 1993a. An overview of unsaturated soil behavior.
Geotechnical special publication, 1(1).
Fredlund, D.G., and Rahardjo, H. 1993b. Soil Mechanics for Unsaturated Soils. John Wiley
& Sons.
Fredlund, D.G., Rahardjo, H., and Fredlund, M.D. 2012. Unsaturated Soil Mechanics in
Engineering Practice. John Wiley & Sons, Inc.
Fredlund, D.G., and Xing, A. 1994. Equations for the soil-water characteristic curve.
Canadian Geotechnical Journal, 31(4): 521–532.
114
Fredlund, D.G., Xing, A., and Huang, S. 1994. Predicting the permeability function for
unsaturated soils using the soil-water characteristic curve. Canadian Geotechnical
Journal, 31(4): 533–546.
Gan, J.K.M., Fredlund, D.G., and Rahardjo, H. 1988. Determination of the shear strength
parameters of an unsaturated soil using the direct shear test. Canadian Geotechnical
Journal, 25(3): 500–510.
Garven, E.A., and Vanapalli, S.K. 2006. Evaluation of Empirical Procedures for Predicting
the Shear Strength of Unsaturated Soils. In Unsaturated Soils 2006. American
Society of Civil Engineers. pp. 2570–2592.
GEO-SLOPE International Ltd. 2015. Stability Modeling with SLOPE/W - An
Engineering Methodology. Calgary, Alberta.
Goh, S.G., Rahardjo, H., and Leong, E.C. 2014. Shear Strength of Unsaturated Soils under
Multiple Drying-Wetting Cycles. Journal of Geotechnical and Geoenvironmental
Engineering, 140(2).
Hilf, J. 1956. An Investigation of Pore-water Pressure in Compacted Cohesive Soils. PhD
Dissertation, Technical Memorandum No. 654, U.S. Bureau of Reclamation,
Denver, Co.
Hinze, J. 2005. Use of Trench Boxes for Worker Protection. Journal of Construction
Engineering and Management, 131(4): 494–500.
Jennings, J.E.B., and Burland, J.B. 1962. Limitations to the Use of Effective Stresses in
Partly Saturated Soils. Geotechnique, 12(2): 125–144.
Khalili, N., and Khabbaz, M.H. 1998. A unique relationship of chi for the determination of
the shear strength of unsaturated soils. Geotechnique, 48(5).
115
Krahn, J. 2003. The 2001 R.M. Hardy Lecture: The limits of limit equilibrium analyses.
Canadian Geotechnical Journal, 40: 643–660.
Krahn, J., and Fredlund, D.G. 1983. Variability of the engineering properties of natural soil
deposits. In Fourth International Conference on Applications of Statistics and
Probability in Soil and Structural Engineering. Florence, Italy. pp. 1017–1029.
Manitoba - Workplace Safety & Health Division. 2011. Guide for Excavation Work.
Ministry of Labour Alberta. (2009). Occupational Health and Safety Code.
Mohamed, F., and Vanapalli, S.K. 2006. A semi-empirical approach for the interpretation
of the bearing capacity of unsaturated soils. University of Ottawa.
Morgenstern, N.R., and Price, V.E. 1965. The Analysis of the Stability of General Slip
Surfaces. Geotechnique, 15(1): 79–93.
National Institute of Occupational Safety & Health. 2013. Trenching and Excavation -
Workplace Safety and Health Topic. US Department of Health and Human Services.
New Brunswick. 1991. Occupational Health and Safety Act.
Nishimura, T., Toyota, H., Vanapalli, S.K., and Oh, W.T. 2008. The shear strength
behavior of a silty soil in the residual zone of unsaturation. In 12th International
Conference of International Association for Computer Methods and Advances in
Geomechanics. Goa, India. pp. 2213–2221.
Oh, W.T., and Vanapalli, S.K. 2010. Influence of Rain Infiltration on the Stability of
Compacted Soil Slopes. Computers and Geotechnics, 37: 649–657.
Oh, W.T., and Vanapalli, S.K. 2018. Modeling the stress versus settlement behavior of
shallow foundations in unsaturated cohesive soils extending the modified total
stress approach. Soils and Foundations, 58(2): 382–397.
116
Ontario - Infrastructure Health & Safety Association. 2017. Trenching Safety.
Pufahl, D.E., Fredlund, D.G., and Rahardjo, H. 1983. Lateral earth pressures in expansive
clay soils. Canadian Geotechnical Journal, 20(2): 228–241.
Rahardjo, H., Heng, O.B., and Choon, L.E. 2004. Shear strength of a compacted residual
soil from consolidated drained and constant water content triaxial tests. Canadian
Geotechnical Journal, 41(3): 421–436.
Rankine, W.J.M. 1857. On the Stability of Loose Earth. Philosophical Transactions of the
Royal Society of London, 147: 9–27.
Richard, A., Oh, W.T., and Brennan, G. 2017. Critical height of an unsupported vertical
trench in an unsaturated sand. In 70th Canadian Geotechnical Conference. Ottawa,
Canada.
Slipp, B., and Gulf Operators. 2018. Construction costs associated with trenching.
Suruda, A., Smith, G., and Baker, S. 1988. Deaths from Trench Cave-in in the Construction
Industry. Journal of Occupational and Environmental Medicine, 30(7): 552–555.
Symons, I.F., Chard, B., and Carder, D.R. 1982. Ground movements caused by deep trench
construction. In Restoration of Sewerage Systems. Thomas Telford Ltd., London.
Terzaghi, K. 1936. The shearing resistance of saturated soils and the angle between the
planes of shear. First international conference on soil Mechanics, 1936, 1: 54–59.
Thompson, L.J., and Tanenbaum, R.J. 1977. Survey of Construction Related Trench Cave-
Ins. Journal of the Construction Division, 103(3): 501–512.
Thu, T.M., Rahardjo, H., and Leong, E.-C. 2006. Shear Strength and Pore-Water Pressure
Characteristics during Constant Water Content Triaxial Tests. Journal of
Geotechnical and Geoenvironmental Engineering, 132(3): 411–419.
117
Trench Shoring Company. 2016. Trench & Manhole Boxes.
Vanapalli, S.K. 2009. Shear strength of unsaturated soils and its applications in
geotechnical engineering practice. In 4th Asia Pacific Conference on Unsaturated
Soils. Newcastle, Australia. pp. 579–598.
Vanapalli, S.K., and Fredlund, D.G. 2000. Comparison of Different Procedures to Predict
Unsaturated Soil Shear Strength. In Advances in Unsaturated Geotechnics.
American Society of Civil Engineers. pp. 195–209.
Vanapalli, S.K., Fredlund, D.G., and Pufahl, D.E. 1997. Saturated-unsaturated shear
strength and hydraulic conductivity behavior of a compacted glacial till. In 50th
Canadian Geotechnical Conference. Ottawa, Canada. pp. 625–632.
Vanapalli, S.K., Fredlund, D.G., and Pufahl, D.E. 1999. The influence of soil structure and
stress history on the soil-water characteristics of a compacted till. 49(2): 143–159.
Vanapalli, S.K., Fredlund, D.G., Pufahl, D.E., and Clifton, A.W. 1996. Model for the
prediction of shear strength with respect to soil suction. Canadian Geotechnical
Journal, 33(3): 379–392.
Vanapalli, S.K., and Oh, W.T. 2012. Stability Analysis of Unsupported Vertical Trenches
in Unsaturated Soils. In GeoCongress 2012. American Society of Civil Engineers.
pp. 2502–2511.
Whenham, V., Vos, M.D., Legrand, C., Charlier, R., Maertens, J., and Verbrugge, J.-C.
2007. Influence of Soil Suction on Trench Stability. In Experimental Unsaturated
Soil Mechanics. Springer Berlin Heidelberg. pp. 495–501.
118
White, F. 2008. National Institute for Occupational Safety and Health (NIOSH) Prevention
through Design (PtD) Workshop Closing Remarks. Journal of Safety Research,
39(2): 203–204.
WorkSafeNB. 2017. Number of trench-related fatalities in New Brunswick.
Yang, H., Rahardjo, H., Leong, E.-C., and Fredlund, D.G. 2004. Factors affecting drying
and wetting soil-water characteristic curves of sandy soils. Canadian Geotechnical
Journal, 41(5): 908–920.
CURRICULUM VITAE
Candidate’s full name: Adin Alexander Richard
Universities attended: B.Sc.E. (Civil), University of New Brunswick, 2016
Journal publications: N/A
Conference presentations: “Critical height of an unsupported vertical trench in an
unsaturated sand”. In Proceedings of the 70th Canadian
Geotechnical Conference. Ottawa, Canada. (2017).